a‘rt-Lraka't‘" r
Ag. ~03 ‘51-). in,
L qr" IFLrngJflr

k-au

;
41:33.»
,‘: u\
L, . .

at: ‘ ~
..

f ' 3‘ . u '
g .. z. ‘ q. . a
{9.5.1 . ‘ A - J. , L u ‘. . .
~ wag . ~ \ . . a»: . .. . . . ~ 3m, .
4511,).‘gxzrmfi ' 3‘33. l‘ - ' 4" ’ . “H "‘ 3" . ' ~ 3.511,”:fi5 1“

-- .1 y: b ‘ . 1 ~ . .1 .' - - a: 3 ‘~

~ «5‘ - w m z;
i u :- .

11W
.4 ' 72’ .-

r . “-
3% ‘ Ai§1x€ m L
a.“ 311': fig?" .

‘ r3»). '2 .
inf} tab-v

‘
9 1 ‘
h;~ V‘N'V
‘1;
741er ‘5
3.5:; ‘ ”3
in w.
. Int;
‘ i

:54!“

g; rzgfis.
0 v ‘
"r.- . @153",
@“tfl ' -
. ‘~ In“ . ‘ . . ‘- ‘32-;
- £32m (hm. . k ‘ w- . u. ,
‘ 4:35, _..~. : .73.; 1.: -‘- v “yr. .
._ .01" n“. ‘ .9, w .0 . xx- ‘5? .,Ei~’1_§:.;§.‘
' K

.
I.
C
r
,, x
, v.
ӣ22.
raw:
v

V

.

, .
0."?
., .

9

4-
3.»
'0

..

‘1

Y l,."'% In}. ) I i I. Y " M1
~'«*I""‘-"- J ‘ " ‘ -" ”"179 '~- £11-...EEH'
t"-"uf="7’s§304 .12; . as ‘ -"*y¢§*§i -.:~~.- 'fitwmrfi‘“

 

E3;

@2355 .
32$ $3354?
.g; ,5. W

1'

ifi‘é‘.

'.’
c

J».
,.

‘ Jam’s

Z . 2‘". T

PM ’-.

'9'“ "
.

TATE UNIVERSITY LIBRAR'ES

Niiiiiii!iiilihufliW m 1 um \HI

3 1293 01020 4885

 

 

 

 

 

 

 

This is to certify that the

dissertation entitled

PARENTAL VALUES AND ELEMENTARY SCHOOL
STUDENT ABSENTEEISM

presented by

Stephen R. Hecker

has been accepted towards fulfillment
of the requirementsfor

Ph.D. degree in Educationai Administration

    

A
Ml!

 

4M a

Major professg'

 

Date February 25, 1994

MSU is an Affirmative Action/Equal Opportunity Institution 0- 12771

 

 

LIBRARY

MICh'Qan State
University

 

 

 

PLACE N RETURN aoxmmwowmmm yuxncord.
TOAVOID FINES rotunonorbdmddodm.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU chn Nflmdlvo Wm»! Opportunity Imam
Wanna-u

PARENTAL VALUES AND ELEMENTARY SCHOOL

STUDENT ABSENTEEISM

By

Stephen Richard Hecker

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Educational Administration

1 994

ABSTRACT
PARENTAL VALUES AND ELEMENTARY SCHOOL
STUDENT ABSENTEEISM
BY

Stephen Richard Hecker

A small proportion of students account for a large proportion of the overall days
of absence from school. Of this small proportion, many are absent for no verifiable
reason. The purpose of the study was to identity which parental values, if any, were
related to the school-absence behavior of students in kindergarten through grade three.

The study was conducted in a midwestern city with a population of just over
100,000. The Fiokeach Value Survey was mailed to 226 homes randomly selected from
among 1,604 homes in which there was both a mother without a high school diploma and
a child in kindergarten through third grade. Ninety families returned the survey.

The Spearman rank correlation coefficient was used to test for relationships
between parental value rankings and days of student absence. The KruskaI-Wallis and
Mann-Whitney tests were used to uncover differences in the way parents ranked values.

Parents who ranked more highly values related to immediate gratification (a
comfortable life, an exciting life, and pleasure) had children who missed more school.
Parents ranking more highly values related to delayed gratification (a sense of

accomplishment, wisdom, and intellectual) had children who missed considerably less

Stephen Richard Hecker

schooL

The major implication of these findings is that improving the attendance patterns
of students exhibiting extreme absenteeism in their earliest school years requires
changing the value systems of their parents. Parents of high-absence students tend to
have value systems which are at odds with schools and social norms. Very young
students who are only marginally socialized in school as a result of absenteeism are thus
being raised by parents who are very different from the mainstream. And although not
all high school dropouts beganitheir school careers as high-absence students, most early
high-absentees eventually drop out. The consequences of early school absenteeism are

serious, both for the student and society.

ACKNOWLEDGMENTS

This research would not have been possible without the support and
counsel of the committee members. Drs. James Costar, Phil Cusick, Dan
Kruger, and George Sherman each provided specific suggestions throughout
the project. Committee chairman Dr. Fred Ignatovich was particularly helpful in
the planning of the study and in providing critical analysis throughout the
research and writing phases. Sue Cooley edited the dissertation and added
clarity, organization, and polish to the final edition. I am solely responsible for
any errors or omissions, but am indebted to each of these individuals for their
significant contributions.

I am also fortunate to have had strong family support. Parents Dr.
Stanley and Mary Hecker and in-Iaws Joseph and Mary Azzarello are ardent
advocates of education for all of their children, as evidenced by the 20 other
college degrees held by their nine children and five spouses of their children.
This strong valuing of schooling was inherited by Linda (Azzarello) Hecker, who
as a wife and professional colleague was instrumental in allowing this research
to come to a conclusion. I am both appreciative of and humbled by the

contributions made by my family.

TABLE OF CONTENTS

Page
LIST OF TABLES ............................................... vii
LIST OF FIGURES ............................................... ix
Chapter
I. INTRODUCTION TO THE STUDY ........................ 1
Introduction .......................................... 1
Absenteeism ......................................... 2
Theoretical Background of the Study ...................... 8
The 12-Value Model ............................. 17
Statement of the Problem ......................... 24
Purpose of the Study ............................ 24
Research Questions ............................. 25
Research Hypotheses ........................... 25
Limitations .................................... 28
Organization of Subsequent Chapters ............... 29
ll. REVIEW OF RELATED LITERATURE .................... 30
Introduction ......................................... 30
Absenteeism ........................................ 31
The Prevalence and Frequency of Absenteeism ....... 31
Causes of Absenteeism .......................... 36
Effects of Absenteeism Among Elementary School
Students .................................... 41
Values ............................................. 51
Cultural Conflict ................................ 51
Values as Antecedents to Behavior ................. 55
Values as Distinguished From Other Concepts ........ 60
The Assessment of Values ............................. 62
The Rokeach Value Survey ....................... 63

V

Summary ........................................... 70

III. ‘ METHODOLOGY .................................... 72
Introduction ......................................... 72
The Setting for the Study .............................. 72
Pilot Study .......................................... 73
The Population and Sample ............................ 74
Absence Rate .................................. 76
Excused Versus Unexcused Absences .............. 82
Survey Techniques ................................... 86
Data-Analysis Techniques .............................. 93
IV. RESEARCH FINDINGS ............................... 96
Introduction ......................................... 96
Initial Findings ....................................... 96
Hypothesis 1 ....................................... 108
Hypothesis 2 ....................................... 115
Hypothesis 3 ....................................... 129
Summary .......................................... 137
V. SUMMARY, IMPLICATIONS OF THE FINDINGS.
RECOMMENDATIONS FOR FURTHER RESEARCH,
AND CONCLUSION ................................. 139
Summary .......................................... 139
Review and Discussion of the Major Findings .............. 141
Implications of the Findings ............................ 152
Recommendations for Further Research ................. 159
Conclusion ........................................ 161
APPENDICES ....................................................
A. Rokeach’s Instrumental and Terminal Value Medians
and Composite Rank Orders for Groups Varying in
Education ......................................... 164
B. The Rokeach Value Survey ............................ 166
C. Cover Letter ....................................... 170
BIBLIOGRAPHY ............................................... 171

vi

Table

1.1

1.2

3.1

3.2

3.3

3.4

3.5

3.6

3.7

3.8

3.9

3.10

4.1

4.2

LIST OF TABLES

Page
Rokeach’s Two-Value Model of Political Ideology ............ 22
A 12-Value Model of Educational Predisposition ............ 23
Definitions of Absence Categories, Based on 180 Days
of Possible Attendance ................................ 78
Distribution of the Population by Absence Category:
Grades 2 Through 4 .................................. 78
Distribution of the Population by Absence Category:
Grades K Through 3 .................................. 80
Distribution of the Sample by Absence Category ............ 82
Residence Changes and Telephone Problems Within the
Three Absence Groups ................................ 88
Surveys Sent and Returned ............................ 90
Ethnicity of the Population .............................. 91
Ethnicity of the Sample ................................ 91
Ethnicity of the Returns ................................ 92
Combined Ethnicity Analysis ............................ 93
Means and Composite Rankings: Terminal Values for
Total Sample ........................................ 98
Means and Composite Rankings: Instrumental Values for
Total Sample ....................................... 100

vii

4.3

4.4

4.5

4.6

4.7

4.8

4.9

4.10

4.11

4.12

4.13

4.14

4.15

4.16

4.17

4.18

A1.

Years of Schooling (Lowest of Seven Categories) of
Rokeach’s National Sample ........................... 101

Years of Education of the Current Sample, by Absence

Group ............................................ 101
Terminal Value Composite Ranking Comparison ........... 103
Instrumental Value Composite Ranking Comparison ........ 104

Correlation of Selected Rokeach Groups With the
Current Sample in Ranking Terminal Values .............. 106

Correlation of Selected Rokeach Groups With the
Current Sample in Ranking Instrumental Values ........... 107

Correlation of Terminal Value Rankings With
Attendance Rates ................................... 109

Terminal Values Correlated With Student Attendance ....... 111

Correlation of Instrumental Value Rankings With

Attendance Rates ................................... 113
Instrumental Values Correlated With Student Attendance . . . . 114
Results of the Kruskal-Wallis ANOVA: Terminal Values ..... 116
Results of the Kruskal-Wallis ANOVA: Instrumental

Values ............................................ 117
Results of the Mann-Whitney Test: Terminal Values ........ 124
Results of the Mann-Whitney Test: Instrumental Values ..... 125
Summary of Results--Terminal Values ................... 130
Summary of Results-Instrumental Values ................ 134

Instrumental Value Medians and Composite Rank Orders
for Groups Varying in Education ........................ 164

Terminal Value Medians and Composite Rank Orders for
Groups Varying in Education .......................... 165

viii

LIST OF FIGURES

Figure Page
4.1 Parental Mean Rankings, by Absence Groups, for

Terminal Values Ranked Differently as Tested

by the Kruskal-Wallis ................................. 119
4.2 Parental Mean Rankings, by Absence Groups, for

Instrumental Values Ranked Differently as Tested

by the KruskaI-Wallis ................................. 123

CHAPTER I

INTRODUCTION TO THE STUDY

InlLQdumiQn

The major themes of the philosophy of public education, ranging from
providing for an enlightened citizenry to supplying a productive work force, are
based on afundamental assumption: School will benefit children. However, if
children are not in school, the institution is unlikely to benefit them or society.
The hopes and dreams attached to education are irrelevant for absent children.

This study is an investigation into one reason for absenteeism among
urban elementary school students. The problem is that a small proportion of
students account for a large proportion of the overall number of days of absence.
The researcher’s premise was that parental values have a significant influence
on this phenomenon. This chapter contains a discussion of the problem of
absenteeism, the theoretical background of the study, a statement ofthe problem
and purpose of the study, the research questions and hypotheses tested in the

study, and limitations of the research.

2
Absenteeism

For too many young children, classroom life is not a daily occurrence.
Their attendance is occasional, even though they are "regularly enrolled."
Teachers are particularly aware of this problem. They often cite the frustration
of trying to help a student who frequently misses school. School districts are
aware of the problem, as many employ attendance officers to encourage
students to be in school. Absenteeism cases have been brought to local courts.
States have enacted compulsory attendance laws, and the Supreme Court has
adjudicated cases relating to the problem. In Goss v. Lopez (1975), the Court
even prescribed a number of days (10) beyond which school absence (in this
case due to suspension) was excessive.

There is evidence that frequent absence can be detrimental socially,
economically, and educationally. Many researchers have found a relationship
between truancy and delinquency (Dawson, 1896; Healy, 1915; Reid, 1985).
Robins and Ratcliff (1978) reported that school absenteeism is the childhood
symptom most reliably associated with an increased rate of adult deviant
behavior. Similar conclusions were reached by Ferguson (1952), Hersov (1960),
Stott (1966). Tennent (1971), May (1975), and Farrington (1980). Robins and
Ratcliff also reported that absenteeism in grade school "did powerfully predict
truancy in high school" (p. 70), and Galloway (1980) also found that the two were

significantly related.

3

Truancy in secondary school, in turn, influences graduation rates "greatly"
(Robins 8 Ratcliff, 1978, p. 73), which has a substantial subsequent economic
effect. Adams (1978) conducted a seven-year study in which he found a positive
relationship between years of school completed and the number of weeks
worked per year and hourly wages. Duncan and Hoffman (1991) found severe
economic consequences of dropping out for females. Jencks (1991) reported
that jobless poverty increased for male dropouts between 1967 and 1987.

Other researchers have described the academic consequences of school
absence. Jencks (1972), Fogelman (1978), and Sinclair and Ghory (1987) are
among the investigators who found strong positive relationships between school
attendance and academic achievement.

Critical to the study of absenteeism is the definition of the term itself. In
W (1988), absenteeism is defined as "absence
from work, school, etc., especially when deliberate or habitual (p. 5, emphasis
added). This is contrasted with the definition, but particularly with the etymology,
of the word "absence." Its origins are traced to Middle English, then Old French,
then to the classical Latin root absens, which is the present participle of the verb
abesse. This word combines ab-, meaning "away,“ and esse, meaning “to be"
(WW, 1988, p. 5). It is important to note this
difference between the words "absence" and ”absenteeism.“ The former refers
to a state of being, whereas the latter introduces the notions of recurrence and

purposefulness. This distinction is central to an understanding of the problem.

4

A third term that is important to this study is "truant" The etymology of
this word is also Middle English and Old French, with a meaning of beggar and
“vagabond.” The root is Celtic and Irish, with a meaning of "wretched." The first
definition of "truant" given in Webster’s is "a lazy, idle person (now obsolete)"
The second is "a pupil who stays away from school without permission." The
third is "a person who neglects his or her work or duties" (p. 1434).

Application of these definitions promotes a greater degree of
understanding in the study of the phenomenon of students missing school.
Absence can be viewed as simply the state of not being in school. It does not
connote habituality or deliberateness. Absenteeism involves some measure of
willfulness or premeditation, and it adds the idea of repetition. Truancy involves
a more negative social connotation (beggar, lazy, wretched) than the other two
words and suggests culpability (pupil who stays away, person who neglects).

In general, the word “truancy" is used to describe older students who are
not in school for long periods of time or who are absent for many short periods,
without a verifiable excuse. Older students are more often away from school
without parental knowledge or permission than are younger students. Also, the
definition and etymology of the word are more negative in tone, which agrees
with the generally more immediate, if not more serious, consequences of missing
school as an older student (Gray, Smith, & Rutter, 1980).

Truancy, then, describes students who stay away from school of their own

volition. Older students (ages 10 and older) who often are not in school

5

frequently make this decision themselves, and commonly their parents do not
know that they are missing school. Younger students (under 10 years old)
typically do not absent themselves from school; when younger students miss
school, their parents very often know when the absence is occurring. It is hard
to imagine a second-grade student setting off for school and then going
elsewhere and "hanging out," yet that scenario is common among high school
students. This is due, in part, to the fact that younger children generally are more
closely supervised by their parents. Also, elementary schools typically contact
students’ homes the day of an absence, whereas secondary schools mail
absence reports with grades every six or eight weeks. In addition, younger
students are not influenced by their peers to the extent that older students are.
Therefore, to examine more directly the relationship between parental
values and students” absenteeism, this study involved students in the five- to
eight-year-old range, or kindergarten, first, second, and third graders. The focus
was on the issue of absenteeism, not on the nature of truancy, because few
young children can be said to be truant, given the aforementioned definition.
The issue of meaning, particularly when discussing children who are not
in school for long periods of time, is problematic in the literature. Although many
reasons for school absence exist, two distinct patterns have been reported
among younger children. The first is absence involving verifiable excuses, such
as bona fide medical illness including school phobia and other events such as

family vacations or out-of-town trips by families for business, funerals, and so on.

6
This pattern is marked by parental efforts toward getting the child back in school
or maintaining the continuity of the child’s school-related activity, and usually
both. This pattern, even when it results in absence for extended periods of time,
is most closely related to the definition of "absence" discussed above, in that the
child simply is not in school.

The second pattern of absence involves unverifiable excuses and a lack
of parental effort to return the child to school or to support academically related
work for the child. Terms such as "avoidance of school" (Klein, 1945, p. 263),
"inappropriate ’home—bound’ school absence" (Waller 8 Eisenberg, 1980, p.
210), "voluntary withholding by a parent" (Galloway, 1980, p. 150), "abscondings"
(Clarke, 1980, p. 111), and "fail to attend school regularly without adequate
reason" (Berg, 1980, p. 137) characterize the difficulty researchers have had in
agreeing on a single definition for this second pattern. The pattern conforms to
the definition of "absenteeism" as previously detailed, so the term is used herein
to represent this model of school absence.

Regarding the frequency of absence, the Goss v. Lopez decision
suggests that an absence rate of more than 5.5% is excessive. Cherry (1976)
determined ”poor attendance" at the 7% absence rate. The Scottish Education
Department (1977) defined "prolonged truancy” as more than one month over the
course of a school year, an absence rate of 12%. Robins and Ratcliff (1980)
defined ”high absence“ as missing more than 20% of the school days within a

ten-week period three times in a school year. They explained, "We are including

7

only those youths with rates of absence so high that they are unlikely to be
explained by illness or parents’ keeping the child at home" (p. 68). Farrington
(1980) set the threshold of high absence at 21.3%. Galloway (1976) cited 50%
absence rates as excessive. Berg (1980) reported on attendance cases coming
to courts involving children missing 75% of possible school days. Although this
range in excessive-absence rates (5.5% to 75%) is large, in the bulk of studies
on absenteeism, a rate of about 20% has been cited as the point of excess.

The behavior of particular interest in the current research is absenteeism
coupled with unverifiable excuses, which suggests a sociological rather than a
psychological or medical inquiry. Other studies into these sociological cases
have been undertaken partly because of the known negative consequences of
excessive absences for the child and partly because of a more general interest
in the characteristics of poverty, delinquency. deviance, and so-called family
dysfunction. The Children’s Defense Fund (1974), Galloway (1982), and
Walberg, Bakalis, Bast, and Baer (1988) are among those who have found
strong relationships between absenteeism and social, economic, family structure,
and other family background factors. Poverty, racial and class differences, and
cultural conflict frequently have been invoked in discussions of absenteeism.
The reported correlations tempt one to assume that these conditions cause high
absenteeism. However, there are poor children, hungry children, children of
physically abusive families, children ofsubstance-abusing families, and children

of broken homes who do manage to get to school regularly. Clearly, some

8

parents, even in the face of difficult life circumstances, send their children to
school regularly. Others do not. The current research was undertaken to
investigate the relationship between absenteeism and a specific set of parental

differences (parental values) rather than other possible causes.

W

The fact of absenteeism invites an investigation into its cause. It would
appear that parents would want their children to be in school. There are negative
social, educational, legal, and economic consequences for missing school.
Likewise, there are relatively strong societal norms censuring school
absenteeism, for both the child and the parent. In addition, it is stressful just
having children around the house when they should be at school. Why would
parents allow their children to stay home from school so often, in the face of
seemingly compelling reasons supporting school attendance?

Deviant behavior, indeed all behavior, has been viewed as having been
caused by society, the individual, the gods, and combinations of these and other
inducements. For centuries, the issues of free will and determinism have fueled
religious, economic, sociological, psychological, philosophical, and many other
debates. Thetheories of Tbnnies, Durkheim, Parsons. and Rokeach underlie the
present research. These writers have provided considerable evidence in support

of general principles upon which the hypotheses formulated for this study were

9

based. Thistheoretical setting provides both a logical/rational background forthe
study and a comparatively direct path through circuitous ideological terrain.

The general theory upon which this study was based is that human
behavior results from a complex interplay of individual, family, and other social
circumstances, including the structure of the society in which the behavior is
observed. Humans are not totally passive, simply responding to stimuli, nor are
they totally active in the sense of conducting their affairs in isolation from external
influences. In the current inquiry into school absenteeism, the writer relied on
this interrelatedness between the individual and society as a fundamental basis
for explaining the behavior. A major component of this symbiosis is the value
pattern of parents andits subsequent effects on children’s behavior. The
selected theorists whose works are cited herein provided support for this line of
thought

Tonnies (1957) described two different social structures, which he called
gemeinschafl‘en and gese/lschafien. Behavior and beliefs are markedly different
under each of these hypothetical orientations. Gemeinschaflen structures
involve small, communal relations that are characterized by face-to-face
interaction. There is an emphasis on collectivism and present (as opposed to
future or past) orientation. The importance of the group, and the secondary role I
of the individual, is evident. There is demonstrable resistance to change. In

general, this structure represents agrarian or village-style communal life.

10

Gese/Ischafien, on the other hand, is a structure characterized by an
acceptance of the value of change. The emphasis is on the furtherance of
individual ends, not collective activity. Relations between people are brief and
are of secondary importance to the pursuit of idiosyncratic goals. Change is
believed to be good. This logical-rational structure emphasizes differentiation
and working toward the future.

Tonnies’s treatment of the concept of human will thus involved the
consideration that the individual and collective aspects (hence the two structural
types) influence one another. The main contribution of his work to the present
study is the introduction of the theory that there is a relationship between social
structure and human behavior.

Durkheim advanced the idea of an interplay between the individual and
society in developing his notion of the "collective conscience." Pitts (1961)
summarized this aspect of Durkheim’s theory:

Society is essentially a set of ideas shared by individuals. Social facts are

things, but things that exist only in the minds of individuals. Society, like

religion, is abstract, normative, and emotional. As an object of
investigation it is influenced by physical facts, size of collectivity,
existential values, complexity of the division of labor, and the
characteristics of individual psychology. but it is not reducible to any one
of these factors. The maintenance of consensus and the maintenance of
order are the organizing principles of Durkheim’s society. . . . Each society
will have the collective representations and the values that it needs in
order to operate in its milieu. Society has an inherent authority. (p. 686)

Later, Pitts elaborated on Durkheim’s position regarding this social

authority: "Society constrains the individual through the attitude of moral respect.

11
This moral respect is not derived from intrinsic properties of the ideas that are
shared, but from the fact that-they are shared" (p. 719). Durkheim (1951) himself
was clear about this point:

As a matter of fact, the empire which [society] holds over consciences is

due much less to the physical supremacy of which it has the privilege than

to the moral authority with which it is invested. If we yield to its orders, it
is not merely because it is strong enough to triumph over our resistance;

it is primarily because it is the object of a venerable respect. (p. 720)

Thus, the search for causality in the study of social problems must include
attention to the relationships among the larger society, the individual, and the
level of the individual's identification with the collective conscience.

Talcott Parsons advanced the theory of the nature and extent of this
interrelatedness. Wu (Parsons 8 Shils, 1951)
and Ihe_S_o_Qiel_S¥S_tem (Parsons, 1951) served to integrate the prevailing
theories on the interaction of culture, social systems, and the personality.
Parsons (1955) developed the concept that four functional imperatives are
associated with all systems of action, and he investigated the process of
socialization. He later described the relations of a social system to all of its
subsystems (Parsons, 1956). Only a fraction of his contributions can be
discussed here, but that portion is fundamental to the theoretical background of
this study.

Parsons (1961) described the four functional imperatives of any social

system as pattern maintenance, integration, goal attainment, and adaptation.

Around these functional imperatives Parsons wove the concepts of role,

12

collectivity, norm, and value. It is at this point that the linkage is established
between the individual (more particularly, the systems [personality, psychology]
within the individual) and the larger social system. Parsons defined values as
shared patterns of normative culture. Roles are normatively regulated behaviors,
and collectivity is the system of interaction of two individuals. Parsons then
stated, "The social systems with which the sociologist normally deals are
complex networks of many different types or categories of roles and collectivities
on many different levels of organization“ (pp. 42-43). Of particular significance
for this study, Parsons wrote that "values define the direction of orientation that
is desirable for the system as a whole" (p. 44).

The imperative of pattern maintenance has one primary component, called
socialization. For Parsons, the primary focus of socialization is the personality.
Early socialization, or development of the personality, is universally
accomplished by a "kinship unit," which always includes the nuclear family. And
although the family may operate at a low level of generality regarding specific
value patterns, it nonetheless has a critical role in the process of socialization

and internalization of values (p. 58).1 Thus, Parsons showed that the relations

 

1Parsons argued that the family in a highly differentiated society has passed
most of its historic kinship structures to nonkinship units, which has weakened
its effect. Formal education supplants kinship units in the process of personality
development in highly differentiated societies. However, it is proposed in the
present study that in such societies some families allow children to stay home
from school, so the kinship unit in this case must necessarily exert an influence
because the child is not in school.

13
among the individual, the family, and the larger society are complex and
interactive at many levels.

Regarding the family and the socialization process specifically. and the
relationship of both to the concept of values, Parsons (1955) wrote:

The central focus of the process of socialization lies in the internalization

of the culture of the society into which the child is born. The most

important part of this culture from this focal point consists in the patterns
of value which in another aspect constitute the institutionalized patterns
of the society. The conditions under which effective socialization can take
place then will include being placed in a social situation where the more
powerful and responsible persons are themselves integrated in the
cultural value system in question, both in that they constitute with the
children an institutional/zed social system, and that the patterns have
previously been internalized in the relevant ways in their own
personalities. The family is clearly in all societies, and no less in our own,

in this sense an institutionalized system. (p. 17)

Thus, the parental pattern of values is controlling in the socialization of the
child. Maximum effectiveness, from an integrative point of view, occurs when the
parental-values pattern reflects the values of the larger society. Especially
significant for the current study, Parsons (1951) noted that "patterns of value-
orientation have been singled out as the most crucial cultural elements in the
organization of systems of action" (p. 159). Action, or behavior, results in large
measure from value patterns, and parents play a key role in transmitting values
as well as in controlling the "action" of their children.

It is within this context of the institutionalizing, socializing, and internalizing

of values that Rokeach introduced his concepts. He defined a value as “an

enduring belief that a specific mode of conduct or end-state of existence is

14

personally or socially preferable to an opposite or converse mode of conduct or
end-state of existence" (Rokeach, 1973, p. 5). Parsons’s definition involved
shared patterns of normative culture. Both his and Rokeach’s definitions relied
on the concept of "valuing" (appraising) social goals or standards as an
antecedent to behavior. This is the core of the theoretical basis for the present
study. Given all of the possible reasons for parents’ not ensuring their children’s
school attendance, and given the complex variety of possible causes, the values
of the parents were investigated as one logical explanation for excessive
absence.

Rokeach made several critical assumptions about values. Consistent with
the line of thought presented here, he wrote, "The antecedents of human values
can be traced to culture, society and its institutions, and personality" (Rokeach,
1973, p. 3). Values come from these elements, and these elements in turn
shape values. Rokeach described how social situations often present the
individual with several competing values, a predicament that can only result in
the modification of the hierarchy of values (p. 6).

Rokeach also assumed that all people possess similar values, but to
different degrees. These values are organized into systems. And although the
number of values is relatively small, the consequences of values will be evident
in all behavior (p. 3). Rokeach reminded the reader of the means-ends
dichotomy, which he related to an individual choosing a particular mode of

conduct or looking toward a particular end-state of existence. This formed the

15
basis for his development of two lists of values—those classified as instrumental,
or relating to modes of behavior, and those seen as terminal, or tied to end-
states.

Terminal values comprise some social and some personal values. Some
are society centered, and others are self-centered. Rokeach argued:

It seems reasonable to anticipate that persons may vary reliably from one

another in the priorities they place on such social [brotherhood or world

peace] and personal [salvation, peace of mind] values; that their attitudes
and behavior will differ from one another depending on whether their

personal or their social values have priority. (p. 8)

There are two kinds of instrumental values. Classified as moral values
and competence values, these are seen as emphasizing either the interpersonal
or the intrapersonal, respectively. An example, Rokeach offered the
competence/moral dilemma. Should one offer intellectual criticism (the
competence value) or be polite (the moral value)?

Rokeach formulated two lists of values, one the terminal values and the
other the instrumental values. He relied on the following logic in performing that
task:

If it is indeed the case that the maintenance, enhancement, and

transmission of values within a culture typically become institutionalized,

then an identification of the major institutions of a society should provide
us with a reasonable point of departure for a comprehensive compilation
and classification of human values. The approach to the measurement
of values described in the next chapter is based, in part, upon an informal

attempt to identify the main values that the various institutions of a society
appear to have specialized in. (p. 25)

16
Specifically, Rokeach’s selection of 18 terminal values resulted from a
review of the literature, a study involving 30 graduate students, and interviews
with a sample of 100 adults in the Midwest. Rokeach reduced the several
hundred values thus collected

. . . on the basis of one or another consideration: We eliminated those

values judged to be more or less synonymous with one another (e.g.,

freedom and liberty, brotherhood of man and equality, peace of mind and

inner harmony), those which were empirically known to be more or less
synonymous (e.g., the correlation between rankings of salvation and unity
with God was over .80), those which overlapped (e.g., religion and
salvation), those which were too specific (e.g., spousehood is more
specific than family security), or those which simply did not represent end-

states of existence (e.g., wisdom is an end-state but education is not). (p.

29)

The 18 instrumental values were selected in a different way. Rokeach
relied on Anderson’s (1968) list of 555 personality-trait words derived from Allport
and Odbert’s (1936) list of 18,000 such words. Because the survey Rokeach
was constructing was intended to be self-attributing, only positive words would
be used. With the elimination of extreme words, sex-linked words,and others,
the Anderson list was reduced to about 200. This number was further reduced
using the following criteria:

- Retaining only one from a group of synonyms.

- Retaining those judged to be minimally correlated

- Retaining those judged to be most important in America.

- Retaining those deemed to be most discriminating across SES, race,
sex, etc.

- Retaining those judged to be meaningful in all cultures.

17

- Avoiding vain or boastful terms (brilliant, etc.). (pp. 29-30)

Rokeach admitted that the process of selecting both lists was intuitive.
His subsequent evaluation of the lists included varimax rotation to determine
factors. Seven factors were found, but they accounted for only 41% of the
variance, with the largest single factor accounting for 8.2%. Thus, he argued that
the intuitive attempt to eliminate duplication was successful. In addition, an initial
12-value form (a total of 24 values combining both lists) was expanded because
it was thought that "too many important values had been left out" (p. 31). The
limit of 1 8 (total of 36) was imposed because more than 18 seemed burdensome
and because the 18 (36 total) selected seemed sufficiently comprehensive (p.

29). Several years’ research were required to select and test the lists of values.

IheJZflLalueJllledel

Rokeach maintained that the value patterns held by all people emphasize
certain values over others. If varying patterns or systems of values exist, and if
such systems underlie behavior, than the task becomes one of identifying the
value patterns that are associated with certain behaviors. Rokeach (1973) and
others have used the Value Survey in just such a manner, investigating civil
rights, religion, politics, honesty, conflict, academic and life-style pursuits,
occupational roles, and other behaviors. "Values are also significantly related to

all kinds of behavior [as well as attitudesl," Rokeach (1973, p. 158) concluded.

18

However, it is not common for a great number of values to correlate with
a specific behavior:

It is possible to specify in advance not all the values that will be

predictably related to a given behavior, but only the main ones. Those

that are the most substantively or logically related to a given behavior
should be the ones that will best predict it. Thus, religious values should
best predict differences in religious behavior, political values should best

predict differences in political behavior, and so on. (p. 122)

Two questions guided this study. First, do systematic or patterned
differences exist among parents relative to their values generally? Next, do
these value-system or pattern differences correlate or associate in any
systematic way with the absence patterns of the children in the family?

To investigate these questions, a logical framework is provided by
Rokeach’s notion that there are "main" values predictably related to the behavior
of interest. In the current study, children’s school attendance was the behavior
of interest, with the parental value pattern considered as controlling. Are there
then “educational values" in the same sense that Rokeach wrote about political
values and religious values? If parents value school, schooling, or education
generally, it seems that they will ensure their children will be schooled. Is there
some subset of the 36 values in the Rokeach Value Survey that is predictably
related to the valuing of education?

Rokeach did not detail a single value or subset of the 36 values in his

survey as relating directly to the valuing of education. It seems logical, though,

that a pattern of "education values" exists, which are held by institutions of

19
education generally (a ”school culture" or collective conscience) and by the
individuals most invested in it. Rokeach (1973) defined an institution as "a social
organization that . . . specializ[es] in the maintenance and enhancement of
selected subsets of values and in their transmission from generation to
generation” (pp. 24-25). Thus, an educational institution supports and promotes
a certain subset of values.

An agreement in values between the institution and the individuals
comprising it would support a longer-lasting relationship between students and
their schools. People who are socialized through regular school attendance
would reflect the norms of schools, norms that these people in turn shaped.
People who spend little time in school likely are not well socialized by the
experience, and they probably develop different, if not opposing value patterns.
People holding these alternative value patterns, ones that conflict with the
education pattern, would likely sever their relationships with schools or limit them
as much as possible.

This is not to say that people who leave school early fall to value
schooling. Nor does the researcher intend to engage in the debate regarding the
difference between schooling and education. The point is that some parents are
adamant about their children’s school attendance precisely because of their own
unexemplary educational experience. Conversely, other parents who spent
many years in school but did not acquire a strong sense that education is

important may not be highly concerned about their children’s school attendance.

20

Still other parents could strongly value education but never send their children
to public schools, choosing private or home schooling instead. The Amish, for
example, shun not only public education but all other mainstream American
institutions as well. Yet the Amish, and other parents without much schooling,
and still others with advanced degrees, regard school as important for their
children. These parents, through whatever mechanism (family, social, or
institutional interaction), have acquired a commitment toward, or support for,
educafion.

A value pattern may exist that underlies this support for education. The
"educational value pattern" might be more prevalent among people with more
years of schooling and less common among people with less schooling, due to
the socialization effects of the educational institution. Are there some values
that, according to Rokeach, discriminate between the less-educated and the well-
educated? If so, do these values, taken together as a model for an educational
value pattern, vary among parents? If that variance exists, can it be linked to the
school attendance patterns of children?

Twenty-five of the 36 values included in Rokeach’s Value Survey
"distinguished significantly" (Rokeach, 1973, p. 63) between groups of well-
educated and less-educated people, the difference between "well" and ”less“
defined by years of school completed (Appendix A). The use of these values (or
some of them) as a model of educational values was guided by the logic

discussed above. One danger of using values held by those with schooling to

21
represent values held by those who think school is important (or, more precisely,
values held by those who think school is important for their children) is that these
might be two distinct groups of people. However, Rokeach did not establish an
educational value pattern, so the best existing model is that suggested by the
value-pattern differences that exist among people with various years of
schooling.

If an educational value pattern is uncovered in this study, it would be
logical then to test the model on various populations, including people who
evidence strong support for schooling in other ways (voting history or
volunteering, for example) but who did not go far in school themselves, against
the years-of-schooling model suggested by Rokeach. The researchers purpose
in the current study, however, was to determine whether variations in parental
value patterns were linked to elementary school students’ attendance patterns.

Just how many of these 25 values ought to be included in a model of
educational values? Rokeach devoted two chapters to the concept of a two-
value model of political ideology. Tapping only two terminal values, freedom and
equality, Rokeach analyzed orientations toward communism, socialism, fascism,
and capitalism. He was confident in the identification of just two words that

differentiated among these ideologies. These relationships are shown in Table

1.1.

22

Table 1.1: Rokeach’s two-value model of political ideology.

 

 

 

 

 

 

 

Freedom--Low Freedomuligh
Equality—Low Facism Capitalism
Equality--High Communism Socialism

 

Although not as parsimonious as Rokeach, the present investigator
proposed a 12-value model of educational predisposition. The model was
constructed from the values Rokeach found to discriminate between well-
educated and less-educated people. The researcher used two guidelines in
choosing the specific values to test in the investigation:

1. The p-value reported by Rokeach as discriminating among varying
levels of education was .001.

2. The rankings relationship is essentially stepwise across the
educational levels.

Of the 25 values Rokeach found to distinguish significantly between the
less-educated and the well-educated, five were significant with p-values of .05
or .01, and they were excluded from the current study because they were not
significant atthe .001 level (nationa/security, self-respect soda/recognition, true
fiiendsh/p, and courageous). Six more were excluded because the rankings did
not rise or fall in an essentially stepwise fashion across the educational levels (a
world at peace, family security, inner harmony, forgiving, helpful, and polite). As

an example of this group, the median rankings reported by Rokeach for family

23
secun'tywere 4.5, 4.6. 3.7, 3.3, 3.5, 3.6, and 6.6 across the seven educational
levels (see Appendix A for the value medians for all values across the seven
educational levels). Two other values (happiness and cheerful) have been
dropped from the instrument since Rokeach’s national survey. The remaining 12
values, six terminal and six instrumental, comprise the proposed model for this

study (see Table 1.2).

Table 1.2: A 12-value model of educational predisposition.

 

 

 

 

 

 

 

 

 

TERMINAL VALUES INSTRUMENTAL VALUES
A comfortable life Clean
A sense of accomplishment Imaginative
Mature lo ve Intellectual
Pleasure Logical
Salvation Obedient
Wisdom Responsible

 

 

The value pattern held by people who are well educated serves as a proxy
for a general school culture, particularly as the pattern contrasts with that held by
people who did not spend much time in school. Parents who do not send their
children to school on a regular basis do not support schooling and thus are likely
to exhibit a value pattern similar to that of parents who did not spend much time
in school. Parents who do send their children to school regularly support

schooling and thus are likely to have value patterns similar to those of the well-

24
educated, even if they (the parents) are not well-educated themselves. Shared
values, which are so important to socialization and cultural integration (Parsons,
1955), do not exist between the school culture and the parents of children who
miss a lot of school. Therefore, school attendance, although powerfully
mandated by the prevailing mainstream culture, becomes optional for some

families. The conflict in values is controlling.

Siatemenleflhefitbblem

A small proportion of students account for a large proportion of the overall
days of absence from school. Poorer students in urban areas, even very young
students, are more likely to be among the highly absent. Yet not all poor
students miss a lot of school. In fact, many students who come from the poorest
families attend school regularly. In reviewing the literature on absenteeism, the
researcher noted few investigations on this problem. Much more common is
research on older students who drop out of school. Thus, the current study was
undertaken to explore the issues surrounding excessive school absenteeism
among young children in order to better explain the phenomenon, focusing on

the values held by the parents as a primary determinant.

Eumcseefibefitudy
The researcher’s primary purpose in this study was to identify which

parental values, if any, are related to the school-absence behavior of children.

25
The findings may provide insights into a problem that persists over time and

across cultures.
Beseambfiuestibns
The following questions were posed to provide a framework for the study

on the relationship between parental values and the school absence of young

children:

1. How did parents in the current sample rank the values on the
survey?

2. How do these results compare to those from other samples?

3. Are these rankings by parents related to the school attendance

patterns of their children?

4. As parents are grouped on the basis of their children’s school
attendance rates, are differences in value rankings evident?

5. How accurately does the 12-value model predict the value-ranking

differences found in the study?

W

The first hypothesis resulted from the question of whether a relationship
existed between the rankings of values by parents and the school attendance

records of their children. The hypothesis was:

Hypothesisjz There is no relationship between parental value rankings
and their children’s school attendance patterns.

26

Hypothesisja: There is a relationship between parental value rankings
and their children’s school attendance patterns.

The second hypothesis was founded on the question of whether
differences existed in the rankings parents ascribed to the values on the survey
when the parents were grouped according to their children’s school attendance
patterns. The hypothesis was:

Hypothesm: When grouped on the basis of their children’s school
attendance patterns, there are no differences in parental value rankings.

Hybetheeisia: When grouped on the basis of their children’s school
attendance patterns, there are differences in parental value rankings.

The third hypothesis was proposed as a result of the logic of the
construction ofthe 12-value model of educational predisposition. The researcher
hypothesized that the value patterns of parents whose children have varying
patterns of attendance would vary in a manner similar to the value differences
found in people having different levels of education. The 12—value model
describes these differences. The researcher hypothesized that parents who sent
their children to school on a regular basis (low-absence parents) would rank
more highly the values a sense of accomplishment, wisdom, imaginative,
intellectual, logical, mature lo ve, and responsible). It was also hypothesized that
parents whose children missed a lot of school without substantive excuses (high-
absence parents) would rank these values low. It was thought that the opposite
trends would hold for a comfortable life, pleasure, salvation, clean, and obedient;

high-absence parents would rank these as more important. In both cases, it was

27

sensible that parents whose children had absence patterns that were neither high
nor low would rank these values-between the rankings of the high- and low-
absence parents. It was thought that two distinct value patterns would emerge,
one for the parents of high-absence students and a different one for the parents
of low-absence students, as these are extreme groups. As parental value
patterns varied, so too would student absence.

The hypothesis suggested by the proposed relationship among the groups
of parents relative to the 12—value model was:

Hypothesis}: The rankings of low-absence parents (Rm) for a sense of

accomplishment, wisdom, imaginative, intellectual, logical, mature lo ve,

responsible, a comfortable life, pleasure, salvation, clean, and obedient

(educational values) will be the same as those of the medium-absence

(RM) and high absence (ngh) parents:

Rm for all educational values = RMed = RHigh

Hypothesisja: The rankings of a sense of accomplishment, wisdom,

imaginative, intellectual, logical, mature lo ve, and responsible (educational

values set 1) will be highest for the low-absence parents, less high for the

medium-absence parents, and lowest for the high-absence parents:

Rm, for educational values set 1 > RMed and > RHigh

mm: The rankings of a comfortable life, pleasure, salvation,

clean, and obedient(educational values set 2) will be highest for the high-

absence parents, less high for the medium-absence parents, and lowest

for the low-absence parents:

RHigh for educational values set 2 > RMed and > RHigh

28
l . 'l l'

The concerns for the limitations of a study include generalizability,
reliability, and validity. Each is discussed briefly in the following paragraphs.

The design of this study controlled for known differences in value patterns
across income levels by limiting the sample to families in which it was indicated
that the mother had not completed high school. By intentionally truncating the
income range by this proxy, the confidence to infer that results might apply to
other income groups was reduced. Similarly, the sample was limited to families
with children in kindergarten through grade 3 so as to ensure that truancy was
not involved as a reason for school absence. This restricts the opportunity to
generalize any findings to populations involving older children. In addition, as a
result of time and financial limitations, the sample involved families from one
urban school district at one time (one school year). This reduces the certainty
with which one can apply any findings to nonurban situations in other time
frames.

Kitwood (cited in Buros, 1978) and Cohen (cited in Euros, 1978) raised the
reliability questions. These involved the ipsative nature of the RVS and the test-
retest correlations reported by Rokeach. The reliability problem presented by the
ipsative nature is that one cannot compare the relative strengths of two
respondents’ rankings. Kitwood’s suggestion that a Likert-type scale would solve
this problem is not convincing. He did state that the test-retest reliability of the

RVS is low, but not unduly so.

29

Kitwood (cited in Buros, 1978) and Mueller (1974) also mentioned the
main concern regarding that validity issue, that being the use of the RVS with
respondents whose academic attainment is below average, or who may have
difficulty understanding the terms or directions included in the RVS. The concern
is that the respondent have sufficient reading skills, and powers of judgment, to
understand and complete the survey. However, the writer did investigate the
ability of mothers without high school diplomas to understand the values and the
requirements and directions of the survey directly with a group of these mothers
in a pilot study before administering the instrument. In addition, other
researchers have successfully employed the RVS with other samples, including
respondents without high school diplomas (Carroll, 1973; Feather & Cross, 1975;

Jenkins, 1974; McCarthy, 1972; Toler, 1974).

QwenizatieneLSubsecuentfibaptens

Chapter II contains a review of the literature on school absenteeism and
school dropouts, the argument for the 12-value model, and an extensive review
of the theoretical background for the study. The methodology of the study is
explained in Chapter III. Results of the data analyses are included in Chapter IV.
The results are discussed as they relate to the hypotheses formulated for the
study. Chapter V contains the conclusions drawn from the study findings,
practical and theoretical implications, and the researcher’s reflections relative to

the study.

CHAPTER II

REVIEW OF RELATED LITERATURE

IntLoduelien

In this study, the researcher was concerned with the relationship between
the school absence behavior of early elementary school students and the value
systems held by the parents of these students. The existence of absenteeism
as an enduring problem across grade levels has been well established. A focus
on early elementary students, though, has been comparatively rare. Research
on absenteeism generally has centered on older students and the problems of
truancy and dropping out of school. Yet there are indications that early
absenteeism is related to truancy and dropping out. Furthermore, there is a logic
that connects the three, as shown in this review. Thus, the first section of the
literature review is focused on absenteeism. Evidence of the prevalence and
frequency of absenteeism, discussion of the causes of absenteeism (with initial
support for the parental-values-based hypothesis as contrasted with other
possible causes), and research into the negative effects of absenteeism as
defined in the present study are provided.

Literature regarding the notion of values is discussed in the second

section. The general concept of cultural conflict is explored, as is the central role

30

31

of values, both within such conflict and as an antecedent to behavior. Also,
literature discussing the differentiation of attitudes, beliefs, and values is
presented.

The work of Milton Rokeach in defining the value-pattern concept and in
designing a measurement instrument is central to the present study. Therefore,
the issues surrounding the measurement of values are examined in this final

section, and the Rokeach Value Survey is reviewed.

Absenteeism

IlaefimalenbeeuiELeeuenexeLAbsemeeism

In the late nineteenth century, Dawson (1896) studied 52 "juvenile
delinquents" (p. 227). The average age of the 26 boys was 15, and that of the
26 girls was 16. Dawson concluded that there were both hereditary and
environmental causes for the delinquency ("delinquent parentage" [p. 248] and
"bad environments" [p. 253]). As part of the description of bad environments,
Dawson referred to the children’s being allowed to "run the streets in idleness"
(p. 254), and he described their "irregular attendance at school" (p. 255).
Although Dawson did not specify the ages at which these behaviors were
observed, he made obvious his overall impression of early childhood for these
youths: "The first elements to be noticed in the early surroundings of these
delinquent children are the poverty and improvidence of the parents" (p. 254).

L. W. Kline (1897), a contemporary of Dawson, wrote an article detailing

the relationship of truancy to the migratory instinct. His primary argument was

32

that excessive school absence was the result of a biological imbalance involving
fluctuations in the metabolic rate. As an organism’s metabolic rate changes and
action is needed, the organism will change its surroundings in lieu of changing
itself. Thus, "running away and truancy are forcible protests against the narrow
and artificial methods of the school room, a rebellion against suppressed activity,
and a denial of free outdoor life" (p. 417). But, Kline noted, "the intimate
correlation between the child’s home and its emotional and ethical life suggest
some very fundamental questions concerning the connection between property
and good citizenship" (p. 418).

Migratory-instinct causation gave way, for Kline, to a social explanation,
touching on the role of the parent and poverty. Later, he detailed the
explanations 9 to 14 year olds gave their teachers regarding school absence.
Among the excuses given before the turn of the century were headache, sick,
mother sick, father in hospital, teeth pulled, measles, cramps, disease in house,
earache, pain in stomach, running errands, kept home to mind the baby, and had
to help wash; the list has a familiar ring to the current practitioner.

More recently, researchers have probed the severity of the problem of
absenteeism in the urbanized postindustrial society, a society that now includes
compulsory attendance laws and significantly higher school attendance rates
than those before the turn of the century. Other changes include a decrease in

manual-labor opportunities and increase in the importance of literacy. School

33
achievement, and its corollary school attendance, became strong mainstream
cultural norms.

As evidence of this trend toward greater school attendance, the
Educational Research Service (1977) reported average daily attendance rates
(ADA) nationwide as improving from 1929-30 to 1975-76. The ADA as a
percentage of enrollment rose from 82.8% in 1929-30 to 92.3% in 1975-76. The
report continued: "In many areas, however, absentee rates average 10-15%,
with the problem most acute in urban schools. Some inner-city schools have
registered absenteeism rates of 30% or more" (p. 3).

In other industrialized cultures, the attendance trends are similar. Berg,
Brown, and Hullin (1988) reported that school attendance in Britain, overall, has
remained at close to 90% for most of the twentieth century; secondary school
attendance in London has remained at about 80% over the same period. Moore
(1966) found significant reluctance to go to school among one-third of a group
of 160 London school children. By age 11, about 20% of this sample still
exhibited reluctance to attend school. Newson and Newson (1977) found that
70 of 700 seven year olds in Nottingham often did not want to go to school, and
they frequently pretended to be sick. The Central Advisory Council for Education
(CACE, 1967) undertook a longitudinal study of a cohort of children born in a
single week in 1958. It was found that 11 year olds were in school between 90%

and 95% of the possible days of attendance, and that there was considerable

34
variation in attendance among children in various areas of the country and
among those in different types of schools.

In examining absenteeism among secondary school students in England
and Wales, Hansard (1974) found that 9.9% of the student body was absent and
that 2.2% (22.7% of those absent) had no legitimate reason for their absence.
In a similar inquiry in Scotland, the Scottish Education Department (1977) found
that 15% of all students had "skipped" school at least once, and that 1.6% of the
boys and 1.1% of the girls had been absent without excuse for at least half of a
six-week period. The Bolton Metropolitan Education Committee (cited in Hersov
and Berg, 1980) discovered that an average of 6.9% of seven year olds were
absent from any half-day session.

Farrington (1980) reported 5.9% of a sample of 411 males as truant in
elementary school. Of particular interest for the present study, he wrote:

The primary school truancy rates of the present sample are about twice

as high asthose found in the National Child Development Study in the Isle

of Wight survey of Rutter et al. (1970), possibly because our children are

urban and working class. (p. 50)

Farrington then mentioned that, in addition to the 24 "truants," 15 other students
had been absent because of lax parental attitudes and another 103 boys had
been absent "because of illness or injury or with no explanatiori' (p. 50, emphasis
added)

In a study of 235 black males in St. Louis, Missouri, Robins and Ratcliff

(1980) found that 45% had been "often truant“ (p. 68), which they defined as

missing more than 20% of school in at least three different quarters of the school

35

year. Reynolds, Jones, St. Leger, and Murgatroyd (1980) studied nine
secondary schools in South Wales and found that attendance rates averaged
80.9% to 84.1% over seven years. Reporting on students aged 10 to 17 who
were assigned to residential educational centers for delinquent behavior, Clarke
(1980) found that 5% of boys and 10% of girls persistently left the facilities
("absconded"). The population in this study averaged 9,000 for each of the 14
years in the period of analysis.

Galloway (1980) wrote about students in Sheffield, England, who missed
more than half of a 6-week period in 1973 and a 14-week period in 1974. For
students aged 5 to 12, the persistent-absence group represented 0.4% of the
total group (N = 5.825) in 1973 and 0.3% of the total group (N = 5,033) in 1974.

Dillon (1949) investigated the school-attendance histories of some 1,200
early school leavers. He found that the elementary school attendance of
students who eventually dropped out "compare[d] favorably with the attendance
of non-school leavers in the average elementary school system" (p. 29).
However, he berated the poor recordkeeping at the elementary level (the
elementary attendance records for 438 of the school leavers were not available)
and pointed out the difficulty of drawing conclusions about the relationship of
attendance to dropping out when such a large chunk of data was unavailable.

The main conclusions regarding absenteeism that can be drawn from the

literature pertinent to the current study are as follows:

36

1. Absenteeism persists across time.
2. Absenteeism persists across cultures.
3. Absenteeism is more prevalent in poor, urban populations than in

more affluent, suburban or rural groups.

CausesoiAbeenteeism

Because absenteeism exists and persists, a search for causes is
appropriate. A substantial number of studies have been conducted on the
truancy, dropping out, and achievement behaviors of secondary school students,
but there has been comparatively little academic inquiry into the problem of
persistent absenteeism among elementary-aged children. Attempts to
understand and explain dropping-out behavior do provide, however, a gateway
to developing an understanding about absenteeism among younger students, so
such efforts are reviewed in this section.

Traditional studies of school dropouts have relied on associating dropping-
out behavior with a constellation of life circumstances. Race, family income,
parents’ level of education, neighborhood composition, whetherthe student lived
with both father and mother, and peer influence are among the conditions that
have been posited as helping to explain and predict dropping-out behavior
(Crane, 1991; Hecker, 1953; Henneghan, 1984; Hollingshead, 1949; Jencks,
1991; McMillan 8 Behrman, 1987; Natriello, McDill, 8 Pallas, 1985; US.

Department of Education, 1984; Wetzel, 1987).

37

Some studies have been undertaken to discover the major ”cause" of
r dropping out behavior, whereas other researchers have attempted to explain
such behavior on the basis of a collection of risk factors. The greater the number
of precipitating or antecedent circumstances, the greater the risk, and the more
likely aberrant behavior (in this case, dropping out) will persist. These analyses
have been useful in describing which factors are closely associated with the
phenomenon, but they have not explained in any substantive detail the
mechanism or process involved in the actual decision to drop out. Unexplained
are the situations of risk-laden students who do not drop out, and virtually risk-
free students who do. Stories of disadvantaged children who eventually ”make
it" and of those others who "had it all" and subsequently failed are key testimony
here.

This method of identifying and assessing a variety of risk factors,
therefore, imperfectly explains dropping-out behavior. Neither can it fully explain
high absence among young children, the juvenile version of dropping out. Three
studies in particular are cited in this section in formulating the logic of the current
study and its reliance on the critical role of the family.

Cervantes (1965) investigated school dropouts and their peer relations,
school experiences, the youth culture, economic-need factors, psychological
tendencies, and the primary group (family). He wrote:

That the enduring core of the "school personality" is primarily fashioned

within the home as the mirror of the specific domestic subculture existent

there is becoming more obvious. The fact that the youth who continues
in school has his origins in a family where personal acceptance,

38

communication, and pleasure are staples is particularly noteworthy. . . .

Our conclusion from our first chapter is that no matter what other variables

are at work, the nuclear family is of critical import in the consideration of

the dropout problem. (pp. 37-38)

Clark (1983) theorized that family life better accounted for student
achievement than did the usually cited risk factors. Although he did not address
the problem of dropping out, Clark described the mechanism of the family’s
contribution to the behavior of the student, a basic tenet of the current study. He
examined students coming from matched backgrounds (i.e., the same high-risk-
factor types of families) with similar cognitive abilities. Of the ten families Clark
studied in detail, five had high-achieving high school seniors and five had low-
achieving seniors. The risk-factor explanation cannot account for these high-
achieving, high-risk students. Clark reported that different expectations, cultures,
values, beliefs, and daily actions of parents better explained the differences in
student achievement than did any other single risk factor or combination of
factors.

In a study of the families of 12 high school dropouts, Okey (1990) explored
the critical role played by family acceptance of dropping—out behavior. He
illuminated the importance of the role of family life and culture as these interact
with school life and culture. Although risk factors play a stage-setting role, in the
end the family’s influence is critical in students’ decisions to drop out.

Galloway’s research (1976, 1980, 1982) in Sheffield, England, extended

this line of thought to the phenomenon of absence among elementary school

39
children. In W (1985). he reported eight
categories of absence for students aged 5 to 11:

- mainly illness

- some illness but with other factors present also,

- absent with parents’ knowledge, consent, and approval,

- parents unable or unwilling to insist on return,

- truancy absence without parents’ knowledge or consent,

- socio-medical reasons (e.g., infestation, scabies. etc),

- school phobia or psychosomatic illness,

- other/could not be rated. (p. 31)

The first and second reasons (primarily illness) accounted for nearly 55%
of the absences of both boys and girls in this age group. The third and fourth
reasons cited (with parent knowledge and parent unable/unwilling to insist on
return, but not medical in nature) accounted for 30% of boys’ absences and 20%
of girls’ absences. The fifth reason, truancy, accounted for less than 5% of boys”
absences and for no girls’ absences. Galloway pointed out that there might have
been school influences on parents’ unwillingness to keep their children in school
(”bullied [by other students] . . . personality clash with teacher” [p. 37] were two
such possibilities). Yet it appears that it was the action of the parent in choosing
to keep the child at home rather than dealing with the problem at school that
contributed to the absenteeism. School personnel may, in fact, be insensitive or
ineffective. But some parents escort their children to school for safety, and talk
with school officials regarding relations with teachers, whereas other parents
simply keep their children at home. The issue is not one of blame, so much as

it is a matter of focusing on the behavior of the parent. After Cervantes, Clark,

and Okey, it is this parental action that is of interest.

40

Galloway was clear in summing up his analysis of school and community
influences, and the idea that family orientation is among the major factors
associated with school attendance among elementary students. However, he
chose to pursue the schools’ contribution:

The research has shown a consistent tendency for absentees to come

from disadvantaged backgrounds. The association between absenteeism

and delinquency is similarly consistent. Yet the evidence is equally clear:

(i) that only a fairly small minority of poor attenders gets into trouble with

the police, and (ii) that a much smaller minority of children from socially

disadvantaged homes are persistent poor attenders. The evidence that
schools themselves exert an important influence over their pupils’
attendance, independent of the catchment area, is not particularly
extensive but all points in the same direction: it would be as myopic to
ignore school influences on attendance as it would be to ignore personal,

social, and family background influences. (pp. 52-53)

Other possible causes of absenteeism among elementary students also
have been studied. Macmillan (1968) investigated socioeconomic factors and
achievement of Spanish-speaking first graders and found a relationship with
attendance. Frerichs (1969) looked into psychosomatic ailments among sixth
graders, noting that the condition was worse if the student came from a broken
home. Bury (1970) found a relationship between air-pollution levels and
attendance among fourth, fifth, and sixth graders in California. Such studies are
examples of efforts to relate certain risk factors with attendance. Although
relationships have been found between a variety of stimuli and the behavior of
interest, the quest for a predictive explanation has not been nearly as well served

by such research as it has by investigations into the role of the family in relation

to school-attendance behavior.

41

EfleflsciAbsenteeismAmeng
EIementarySeboeLStudents

Absenteeism among elementary school students has been reported to
have both short- and long-term effects. The short-tent: consequences of school
absenteeism primarily concern academic achievement, and the long-term
consequences are involved with continuing absenteeism, truancy, delinquency,
and reduced adult income.

Rozelle (1968) reported 24 studies that were conducted from 1923
through 1962. In half of these studies, significant relationships were found
between absence and grades; in the other half, no such relationship was found.
Summers and Wolfe (1975) reported that unexcused absences among
Philadelphia sixth graders (N = 627) had a negative effect on their achievement
as measured by the California Achievement Test and the Comprehensive Test
of Basic Skills. That negative effect was greater for higher-income white
students, for whom five unexcused absences translated into a decline of 2.13
months of growth. For poor students, the five unexcused absences were
associated with a loss of 1.32 months of growth.

Coldiron and Skiffington (1975) found no association between school
attendance and achievement for Pennsylvania fifth graders in the years 1969 to
1973. They suspected such a link, but attributed their findings to the low
variability in attendance at the fifth-grade level (pp. 3637).

Douglas and Ross (1965) reported their work with data from the British

National Survey of Health and Development. School absences were grouped

42
into seven categories on the basis of both frequency and time of absence.
Comparisons were made, controlling for socioeconomic status and cognitive
ability. The researchers concluded that school absenteeism was associated with
declining test scores for students in all social classes below the upper middle
class. Scores could improve if attendance improved, but not for children of the
lower classes. This last finding differs from Summers and Wolfe’s conclusion
that absence had a more negative effect on the higher social classes.
The report by Jencks (1972) on school attendance is pertinent. He wrote:
Taken as a whole, evidence about the effects of school attendance on
test scores is woefully inadequate. Such evidence as there is suggests
that preschooling has few permanent effects, that elementary schooling
is quite important to the development of the skills measured on
standardized tests, and that secondary schooling and college also boost
test scores to some extent. (p. 85)
Jencks examined the effects of wholesale school closings in Holland
during World War II, Prince Edward County in the early 1960s, and New York
City in the fall of 1968 on the students who were excluded from school. He

commented:

These findings imply that if all elementary schools were closed down, so
that growing up became an endless summer, white middle-class children
might still learn much of what they now learn. Some of these children are
taught to read before they enter school anyway, and some of them read
a great deal at home, developing their skills without any help from school.
But most poor black children would probably not learn to read without
schools. The cognitive gap between rich and poor and between black
and white would thus be far greater than it is now. (p. 88)

At the time of his study, Jencks was primarily interested in examining

inequities between blacks and whites in America, so his writing reflects that

43

orientation. Yet he also discovered that within-group disparities were much
greater than between-group differences. Thus, Jencks .might agree that
dropping the words ”black" and "white" from the preceding quotation would not
change the meaning greatly. In other words, school attendance is particularly
important, in an academic-achievement sense, for children from poor families.

The long-term effects of elementary school absenteeism are primarily
residual to the academic losses sustained by absentees, and secondarily result
from the social isolation associated with not being in school. It is difficult,
particularly for children of lower socioeconomic status, to have high academic
achievement while missing a lot of school. Lacking academic skills, students
struggle with increasingly difficult school—day challenges and eventually decide
to drop out before learning the minimal skills. And by missing a lot of school
before eventually severing the relationship altogether, absentees fail to acquire
the mainstream cultural, linguistic, and behavioral habits many employers seek.
Several researchers have provided background and supportforthese assertions,
most notably Robins and Ratcliff (1980).

In their 1968 study of 235 black males born in St. Louis, Missouri, between
1930 and 1934, Robins and Ratcliff reported findings that corroborate the
foregoing argument:

We have shown that elementary school truants accounted forthe excess

mortality in black schoolboys, as compared with expected figures based

on a national white male cohort (Robins, 1968). The excess mortality was

accounted for principally by homicide, but truants also had an excess of

deaths by natural causes. Among those who survived past the age of 25
and so were included in the follow-up, we have shown that elementary

44

school truancy significantly predicted four subsequent childhood events:
being held back in elementary school, dropping out of high school, leaving
the parental home before age 18, and marriage before age 18 (Robins 8
Wish, 1977). Truancy predicted these subsequent childhood behaviours
at a statistically significant level even after we took into account the
number and kinds of other deviant behaviours that preceded the truancy
and so might have explained the subsequent behaviour. We have also
found that men who were truants in elementary school tended to marry
women who truanted in elementary school and that truancy in either
parent was associated with an excess rate of truancy in both sons and
daughters, although transmission of truancy to the sons was more striking
(Robins, Ratcliff, 8 West, 1979). When we compared school records of
the wives and children with those of our index fathers, we found that
elementary school truancy did not predict high school dropout at a
statistically significant level for females in either generation, but that
truancy in the sons of our subjects did predict high school dropout in just
the way that it had for their fathers. This suggested that our finding in the
white child guidance clinic study that truancy is a more potent predictor of
outcome for males than females probably holds in black non-patient
samples as well. (pp. 67-68)

This link between absenteeism in elementary school and problems later
in life was investigated by Robins and Ratcliff in other ways as well. It is
instructive to discuss first the difficulty they had in defining the word "truancy."
They used the criterion of missing more than 20% of school days within three 10-
week periods, believing that "we are including only those youths with rates of
absence so high that they are unlikely to be explained by illness or parents’
keeping the child at home" (p. 68). This last point is contrary to the thesis
presented in the current study, and Robins and Ratcliff were

. . . surprised, therefore, to learn that two-thirds of the children who

qualified as often truant in elementary school had at least one truant

quarter in first grade. Thus, even if absence in first and second grade
does occur more often because of illness or with parental consent than

because of truancy, it seems to establish an attendance pattern that later
is a truant one. (pp. 68-69)

45

Although Robins and RatclifI’s intention was to investigate the results of
excessive school absence, and not the reasons for the absence, these
researchers uncovered a fundamental problem in the framing of questions
surrounding the phenomenon, that of the definition of terms. In the end, as they
noted, early patterns of nonattendance have significant implications for later
attendance patterns. Parentally instigated absence, which for Robins and Ratcliff
was not technically truancy, in fact was difficult to separate operationally. Their
distinction between the two was unnecessary.

Returning to the issue of effects of absenteeism, one of the first results of
truancy, and a rather direct one, is that students who are out of school tend to
stay out of school:

Only 13% of those with no truancy in elementary school became truant in

high school. For those mildly truant in elementary school, the rate of high

school truancy (23%) was almost twice the rate for good elementary
school attenders. Those often truant in elementary school became truant
in 39% of cases in high school, three times the rate for elementary non-
truants. . . . The effect of truancy in elementary school on truancy in high
school is even more striking when we restrict the sample to those whose

truancy was evaluated. For those often truant in elementary school, 49%

were also truant in high school, 3.5 times the rate of those not truant in

elementary school. (p. 69)

However, Robins and Ratcliff reported that "a number of children often
truant in elementary school became good attenders in high school" (p. 70), so
they looked at other deviant behaviors. Being held back in elementary school;

drinking and having sex before age 15; using marijuana, amphetamines,

barbiturates, and/or opiates, or having problems with alcohol before age 18;

46
leaving home; being arrested ; and marrying before age 18 comprised their list of
deviant behaviors. The researchers found that:

Students who were not truant or only mildly truant in elementary school
had fewer of these other deviant behaviours than did those frequently
truant in elementary school. . . . We found . . . that juvenile deviance was
associated with the continuation of truancy from grade school in high
school. Children who were often truant in grade school usually truanted
in high school unlessthey had very little deviant behaviour of other kinds.
Juvenile deviance was also associated with beginning truancy in high
school among those who had not truanted or were only mildly truant in
grade school. Those who did little truanting in grade school rarely began
truanting in high school unless they were extremely deviant in other ways,
in which case almost half became truant. . . . Elementary school truancy
has an impact on high school truancy even when we controlled for other
childhood deviance. Those often truant in grade school were more likely
to truant in high school at each level of our deviance scale . . . [but] we will
need to be careful when we look at the impact of truancy on later
outcomes that we are not simply using truancy as an indicator of the
general level of juvenile deviance. (pp. 70-71)

Given the relationship between early and latertruancy, Robins and Ratcliff
turned to the influence of early school truancy on final educational level, adult
earnings, adult deviant behaviors besides drug abuse, adult drug abuse, and
adult psychological status. Their findings were generally in the expected
direction (that truancy was linked to each), with some unanticipated facets.

Regarding completion of high school, "of those often truant in elementary
school whose truancy continued into high school, 75% failed to graduate” (p. 71).
And, ”while the relationship between high school truancy and dropout seems
inevitable, it is of interest that dropping out of high school was also influenced by

the level of elementary school truancy both among those who did and those who

47
did not truant in high school" (p. 71). Mixing in the level of deviance as
previously defined,

At each level of deviance, the chances of graduating from high school

decreased with an increase in the amount of truancy. The difference is

particularly striking for those with mild amounts of deviance, where more

than 90% graduated without any significant truancy but less than 40%

graduated if they had been truant in high school. . . . [And of the high

deviants] who were not truant in high school, [33% did graduate, but] only

6% did so ifthey were truant in high school, and [none graduated] if they

had been both truant in high school and often truant in elementary school.

Thus truancy greatly influenced high school graduation at all levels of

deviance. (p. 73)

Concerning the matter of adult earnings, Robins and Ratcliff conducted
a multiple regression analysis on the effect of high school truancy, graduation,
and juvenile deviance. They reported that, "with respect to low earnings, high
school truancy was the only important one of the three variables, alone
accounting for 10% of the variance after adjustment for the impact of the other
two variables“ (p. 80). The researchers had excluded consideration of
elementary truancy from this analysis "because high school truancy was a
stronger predictor of adult outcomes than elementary school truancy" (p. 78) and
because it was necessary to simplify the analysis.

Regarding adult deviance, Robins and Ratcliff probed the relationships
among elementary truancy, secondary truancy, and nine deviant behaviors,
including criminality, job problems, violence, and regular illicit drug use. In all
cases, secondary truancy predicted each of these behaviors far better than did

elementary truancy, but this was not true for the deviant behavior of "parental

irresponsibility.” This they defined as either fathering illegitimate children or not

48
supporting legitimate ones. Elementary truancy was much more strongly related
to this adult deviance than it was to any of the others. As expected, those who
were truant in both elementary and secondary school had a higher incidence of
parental irresponsibility than did those who were truantjust in elementary school.
"Parental irresponsibility is the only behaviour more closely associated with
elementary than high school truancy" (p. 76).

In analyzing the effect of truancy on adult drug abuse and adult
psychological dysfunction, Robins and Ratcliff found moderate effects. Drug use
was not predicted by high school truancy,and psychological disturbance was
predicted only slightly. Summing up their findings, the authors wrote:

We have shown that truancy had important implications for both childhood

and adult outcomes of our sample of young black men. Elementary

school truancy. often beginning in first grade, forecast continued truancy
in high school, particularly in those who had other kinds of deviant
behaviour such as early drinking, early sexual activity, illicit drug use, and
delinquency. Boys not truant in elementary school seldom became truant
later unless they began this set of adolescent deviant behaviours at the
same time. Both elementary and high school truancy were associated
with dropping out of school before completing secondary education, and
also with low earnings as an adult. High school truancy was strongly
related to a variety of adult deviant behaviours, and somewhat associated

with psychological disturbance. (pp. 80-81)

In 1978, Farrington (1980) investigated a sample of 411 males in London.
All were about 25 years of age at the time of the study, and longitudinal data
were available since the time the subjects had been about eight years of age.
The link between truancy and delinquency was clear: "Given the overlap

between truancy and delinquency, and the fact that general truancy precedes

convictions, another question which can be asked is whether there are any

49
important differences between truants who become delinquents and truants who
do not" (p. 60). Farrington’s answer was that there is a difference in degree but
not in kind.

The findings from others studies by May (1975), Tennent (1971), Stott
(1966), Hersov (1960), and Ferguson (1952) supported the notion of a
relationship between truancy and delinquency. Healy, in a 1915 work entitled
IhemdeuaLflelinguem, also mentioned such a link.

Regarding the issue of income and years of schooling completed, Adams
(1978) conducted a study for the Upjohn Institute for Employment Research. He
found a significant linear relationship between the two; as years of schooling
increased, so did income. Cherry (1976) found a relationship between poor
school attendance and subsequent frequent job changes as adults, but the job
changing did not result in poorer earnings. Yet Gray, Smith, and Rutter (1980)
summed up the relationship as follows: "We may conclude that truancy,
absenteeism, and school dropout seem to have few effects on employment or
incomes in early adult life, but that there is a substantial association with higher
unemployment and lower social status when older" (p. 347).

It should be noted that the analyses by Adams (1978), Cherry (1976), and
Gray et al. (1980) were conducted on samples ”coming of age” in the 1960s and
19703. Freeman (1991) observed that ”economic opportunities for young male
graduates deteriorated greatly in the 1970’s and 1980’s" (p. 103). Jencks (1991)

noted that ”the [long-term jobless rate] among 25- to 54-year-old men [rose] in

50
both the 19703 and 19803” (p. 94). Thus, in analyzing joblessness in 1959,
1969, and 1979, Jencks wrote, "Jobless poverty increased substantially among
both blacks and whites without high school diplomas" (p. 54). Duncan and
Hoffman (1991), in interpreting national data from the Panel Study of Income
Dynamics, supplied a picture of the same problem for women:

An economic model of teenage behavior should focus on the relative

payoff for good behavior—that is, the difference in likely poverty or median

income between teenagers who did and did not follow the rules. Here the
poverty trends are striking: the difference in poverty rates between the
two groups ofwomen has increased substantially because both black and
white teenagers who dropped out of school or had babies out of wedlock
were much worse off when they reached their mid-twenties in the early

19803 than were such women twenty years earlier. Poverty rates of white

women who either dropped out of high school or had a child out of

wedlock more than tripled and nearly doubled for black women. Median
family incomes at age 25 forthese groups dropped sharply as well. In the

19805 the median income at age 25 for black women who dropped out of

school or had a child out of wedlock was only $8300. (p. 162)

Thus, in both the American and British societies, there are long-term
consequences, often severe ones, of missing school. The relationship between
poor school attendance early in life and subsequent truancy and eventual
dropping out is strong enough to warrant concern. Secondary school truancy
and dropping-out behaviors have been found to be related to a variety of
subsequent adult problems, and these relationships have been well supported
in the literature. Throughout the many investigations that have been conducted
on these topics, the influence of the family often has been citeduhence the

current study into the value patterns of parents as one possible cause for the

school-attendance behavior of young children.

51
Values

One of the most striking observations that has been made concerning the
involvement of school personnel with the parents of young children exhibiting
high absenteeism is the frequency of high levels of parental hostility toward the
school, the representatives of the school, and the world in general. School
secretaries, teachers, principals, school social workers, and truant officers calling
on these families to check on reasons for children’s school absence often
encounter suspicious and evasive if not rude, angry, and verbally abusive
parents. The main complaint of many of these parents is that the school is prying
into the family’s affairs and that the reasons for school absence are not the
school’s business. Indeed, the reasons behind these excessive absences are
often poverty ("We don’t have money for shoes"), family violence ("His daddy
was causin’ trouble here, keepin’ us up all night long"), and other matters that
would be embarrassing to the family if widely known. In this sense, from the
family’s perspective, the school personnel are invasive. And although not all
students exhibiting patterns of high absenteeism have this type of family life,
such conditions are sufficiently prevalent to stimulate an investigation into the

nature of the relationship between parental values and student absenteeism.

CultuLalfibnfllct
The conflict between the school and the families of high-absence students

can be viewed as being lodged in the cultural differences between the family and

the larger society. In Wu (1988), culture is defined

52

as “the ideas, customs, skills, arts, etc. of a people or group that are transferred,
communicated, or passed along” (p. 337). Johnson (1985) wrote that "culture is
a fairly persistent patterned interaction of distinctive behaviors, ideas, and
physical objects” (p. 7). As the conflict with school personnel regarding school
attendance appears to be a ”custom” and "fairly persistent" for many families in
the high-absence subgroup, it is possible that the problem represents a more
general pattern of cultural conflict between the poor (who tend to be
overrepresented in the high-absence category) and the larger society.
Therefore, the literature regarding cultural conflict was reviewed because it
provides one component of the logic of this study. If cultural differences exist
between some families and the school, the school-absenteeism phenomenon
might be better understood through exploring those differences and their general
effect on behavior.

Large-scale immigration into the United States, which ended in the 19203,
gave rise to interest in the problems of people who were raised in one culture
and then lived in another. Mumford (1926) supplied a preliminary definition as
he dealt with one of the problems of the day--that of displaced people seeming
to have no roots:

Unfortunately, a man without a background is not more truly a man: he

has merely lost the scenes and institutions which gave him his proper

shape. If one studies him closely, one will find that he has secretly
arranged another background, made up of shadows that linger in the
memory, or he is uneasy and restless, settles down, moves on, comes

home again, lives on hopeless tomorrows, or sinks back into moumful
yesterdays. (pp. 38-39)

53
Stonequist (1938) wrote about the cultural assimilation of the ”marginal
man“-—one who is “of" one culture but living in another. Although referring to a
different problem (immigration) at a different time, the implications forthe current
study are clear:
In the family and under the influence of the tribe, the sect, or the local
community. . . the individual acquires those habits, sentiments, attitudes,
and other personality traits that characterize him as human. . . . The
marginal man is one whom fate has condemned to live in two societies
and in two, not merely different but antagonistic, cultures. . . . [Conflict is]
an incidental product of a process of acculturation, such as inevitably
ensues when peoples of different cultures and different races come
together to carry on a common life . . . for it is not the mere mixing of
cultures which creates the marginal man, but rather the experience of
conflict of group attitudes flowing from the cultural differences. (p. 88)
Sociopolitical history includes many examples of clearly defined cultural
differences driving significant levels of conflict. Fagan (1984) reviewed the
European exploration and settlement movement from 1488 through the 19005.
In W, he cited "progressive confrontation between an expanding,
sophisticated civilization with radically alien beliefs and dozens of societies that
lived in careful balance with the natural resource of their environment” (p. 5). He
also detailed the carnage and annihilation that took place as Europeans
"discovered” the bounty of the New World.
As an example ofthis conflict phenomenon on the individual level, Seward
(1958) detailed case studies ofclinically defined psychological problems resulting
directly from cultural conflict. Significantly forthe current study she wrote, "Since

the first impact of culture comes through the home, the earliest and most serious

forms of conflict with it are reflected in children’s problems" (p. ix).

54
In W, Riessman (1962) stated, “We view

culture as an effort to cope with the surrounding environment“ (p. 6). Consistent
with the logic of the current study, he developed the notion that there is indeed
a separate culture among the poor, a culture that is different from that of the
larger society. Differences between the two cultures are fertile ground for the
development of a variety of negative consequences, and knowledge ofthe nature
of both the differences and the consequences is suspect: "The understanding
and treatment ofjuvenile delinquents from disadvantaged backgrounds has been
something less than earthshaking, and the cleavage between the deprived child
and the school appears to have reached new heights" (Riessman, 1962, p. xiii).

Some families have cultural orientations that are significantly different from
that of the larger society. Families who experience such differences also may
experience the accompanying conflict. It the family happens also to face
problems of poverty and racial discrimination, whether they are blacks in urban
settings (Gilbert 8 Gay, 1985; Villegas, 1988), Mexican-Americans in border
states (Buenning 8 Tollefson, 1987; Mendelberg, 1986; Trueba 8 Delgado-
Gaitan, 1985), Asians in many parts of the country (Blakely, 19874; Wehrly 8
Nelson, 1987), or Native Americans in the western United States (Chilcott, 1985;
Clark, 1983; Courtney, 1986; Cuch, 1987; James, 1988), the problems are
magnified. The conflict resulting in large measure from cultural differences,

compounded by ethnic/racial and income differences, has thus been well

55

supported in the literature as a continuing problem, and it is advanced here as

one factor contributing to school-attendance behavior.

WW

If, in fact, cultural conflict is controlling, the components of "cultural
orientation” require definition and examination in order to formulate a testable
hypothesis. In nearly all of the works cited in this review of the cultural-conflict
explanation of student-absence patterns, the concepts of behavior, beliefs,
attitudes, and values have been cited as indicants of "culture." Johnson’s (1985)
definition of culture ("a . . . fairly persistent patterned interaction of distinctive
behaviors" [p. 7]), Riesman’s (1952) assertion that "society is . . . instilling a
particular mode of conformity in its members, who then perpetuate the society
as they go about its business, including the rearing of the young" (p. 5), and the
Parsonian concept of pattern-maintenance (see Chapter I) all require certain
indicants of "culture."

To study the conflict associated with suspected cultural variance, one
must direct attention toward behavior, beliefs, attitudes, and values. It is this
path that leads directly to the current study. The behavior is defined (high
absenteeism of young children for no "good" reason). The assessment, then, of
beliefs, attitudes, and/or values of the parents, who control the behavior of the
children, is thus strongly suggested in order to investigate the phenomenon of

variance in school attendance.

56

It is important to note that nojudgment is made here regarding the "worth"
or "goodness" of one cultural perspective over another, .or that there is some
"lack of culture" in any case. The premise is that all behavior is associated with
cultural antecedents, and that understanding the behavior requires an
understanding of that cultural lineage and/or context. Anderson (1991) was
instructive on this point in discussing what he called "neighborhood effects" on
teenage pregnancy. He described the mechanism of cultural effect as follows:

The girls who graduate to the street are products of homes in which they
have relatively little parental supervision and limited family support to
strive for a life much different from the one they are currently living.
Hence, as indicated earlier, decent1 parents often forbid their own
children from regular participation in such groups and label such children
street kids. . . . The girl primarily raises herself with the help of her street-
oriented peers and her mother say[s], "she just grows up." . . . Left largely
to their own devices, these children play with their peers on the street
comers in mild weather, at times until 1:00 or 2:00 in the morning on
school nights. . . . By the age of eleven or twelve many of the street girls
are aware of their bodies and beginning to engage in sexual relations. .
. . [They] become committed to their peer groups, learning to survive by
their wits. . . . Increasingly they "talk to people" (boys), whose sexual
desires begin to mesh with the social needs of the girls. Some members
of this group begin to have babies by the age of fifteen, and soon others
follow. . . . For some, becoming pregnant may be viewed as normal and
as only a matter of time. . . . In time, this primary group can become
something of a family for many of its members who grow up lacking the
emotional supports of one. When this group wins the girl’s allegiance, it
works to shape her dreams, social agenda, values, aspirations, and goals.
At times, it competes strongly with the inner-city family, "decent" or
"street." (pp. 387-388)

 

1Anderson was careful to note that the term "decent" was not his term, but
was defined and used in just this sense by the residents in the black underclass
neighborhoods in which he had conducted ethnographic studies for more than
20 years.

57

The formation of a perspective from which one selects behaviors occurs
in many ways, both in and out of the family. A classroom, a clan, and/or a
neighborhood group can each have an influence.

It is logical, then, to examine the "perspective" from which one operates
when deciding on a particular action. Are action choices based on beliefs about
the world, attitudes regarding choices and consequences, morality and ethics,
or some hierarchy of values? According to the theoretical perspective on which
the current study was based, it is assumed that there is some interaction
between the individual and the larger society regarding action, and that values
play a significant role.

Supporting this central role of values, von Mering (1961) wrote, "Our basic
premise is that knowing and valuing are social facts and, therefore, analyzable
within a scheme of social action" (p. 13). He proposed three types of values,
including existential (cultural), normative (morally directive), and idiosyncratic
(projective) values. He suggested that people operate within the constraints of
a particular grammar of values:

Such a particular hierarchy of "personalized" values is always to be

regarded as a product of an individual’s biographical and cultural situation.

Thus, the individual value profile tends to reflect the cumulative,

preferential trending to certain values and not to others that are

characteristic of the culture as a whole. (p. 243)

Although von Mering maintained that "human value patterns are neither

very simple nor easily predictable" (p. 242), the classification of values he

proposed was successful in helping to differentiate among the five cultures he

58

studied. And, significantly for the logic of the current investigation, his study
"made it possible to demonstrate with reasonable confidence that a high degree
of association between certain value clusters is not a chance occurrence but has
experiential meaning to the valuer" (p. 245). Meanwhile, von Mering
acknowledged that "today, social scientists are conducting extensive and
intensive studies in the formation and role of attitudes, interests, and values in
social perception and action" (p. 5). His work helped establish that values play
a major role in determining social action.

Sigel (1985) edited a volume of articles in which parental beliefs were
linked with variance in a variety of children’s characteristics and behaviors,
including the child’s cognitive level, cognitive development, socioeconomic
status, and mathematics performance. In his own article, Sigel argued for the
centrality of belief. He contended that a belief is a "cognitive representation of
reality" or "a cognitive construct with content" (p. 357). His example of a belief
was the statement, "I believe children learn through experimentation" (p. 357).
"Attitude" introduces the affective component, referring to one’s feelings about
the cognition (belief). "Value," then, refers to "the importance of the intended
action relative to the [actor’s] . . . goals" (p. 357). Although Sigel argued that
"belief" is the core of action, he noted that "beliefs [are] attenuated by attitude
and value" (p. 357). Whether belief, attitude, or value is more fundamental, or

a better predictor of behavior, remains problematic in this definition. The

59
centrality of parental influence is supported, but the causal key remains
undetected.
Kluckhohn and Strodtbeck (1961) provided an argument that addresses
the problem presented by this critical reading of Sigel and is consistent with the
overall theoretical basis for the current study:

Directiveness appears to be causally important in at least two senses.
First, although the several writers cited above have argued strongly and
convincingly that the cognitive and affective elements are inextricably
interrelated, they do not go far in indicating the "why" of this relationship
other than stating that it appears an irrefutable fact that what a people
believe to be true (existential premises) is strongly influenced by their
normative judgments and that contrariwise the normative assumptions as
to what is right and proper are never truly separable from the existential
premises. Selectivity is discussed as an element of the total process, but
it is not clearly defined as an element which is distinctive in having relating
(integrating) and directive (processual guiding) influences upon the other
elements. It is our view that to the extent that the cognitive and affective
aspects of the process are a unity it is because of the directive element,
which is as much, or perhaps even more, biologically given than are the
capacities for either intellection or affectivity.

It is on this basis that we state that in the concept of a value orientation
as a guiding principle, it is the directive element which is of primary
interest. This is the second way in which the element may be considered
as a critically causal one. Any given value system of human beings has
both a content and a direction which derive from biologically given
capacities and predispositions but are not instinct bound, but it is the
directive aspect which is the most crucial for the understanding of both the
integration of the total value system and its continuity through time.

In the realm of observed behavior the integration effect is the thematic
one. All or almost all aspects of the social life of a people give
expression, in varying ways and varying degrees to be sure, to the basic
values which are characteristic of one culture as opposed to another. (p.
9)

60
III III 'I IE on C I

Rokeach (1973) took issue with the concept of value orientation as
advanced by Kluckhohn and Strodtbeck, arguing that their value orientations
were "more aptly described as basic philosophical orientations" (p. 22) in the
motivational-causal path. In support of his own conceptualization, Rokeach
detailed the differences between values and a variety of other concepts,
beginning with attitude:

An attitude differs from a value in that an attitude refers to an organization

of several beliefs around a specific object or situation (Rokeach, 1968a,

1968b). A value, on the other hand, refers to a single belief of a very
specific kind. It concerns a desirable mode of behavior or end-state that
has a transcendental quality to it, guiding actions, attitudes, judgments,
and comparisons across specific objects and situations and beyond

immediate goals to more ultimate goals. (p. 18)

Rokeach mentioned seven respects in which values and attitudes differ.
Values are single beliefs, whereas attitudes refer to several beliefs focused on
one object or situation. Values are not attached to specific situations or objects,
as are attitudes. Values serve as standards, whereas attitudes do not. Values,
because they relate to modes of conduct and end-states, number "only in the
dozens, whereas attitudes number in the thousands" (p[. 18). Values are more
central to a person than attitudes. Value "is a more dynamic concept than
attitude, having a more immediate link to motivation" (pp. 18-19). Last, "the

substantive content of a value may directly concern adjustive, ego defense,

knowledge or self-actualizing functions while the content of an attitude is related

61
to such functions only inferentially“ (p. 19). Rokeach then described differences
between values and social norms, needs, traits, interests, and value orientations.

Rokeach (1980) later clarified his position that values are "shared
prescriptive or proscriptive beliefs about ideal modes of behavior and end-states
of existence that are activated by, yet transcend object and situation" (p. 262).
Indeed, his opening sentence in this article was, "While I have had a long-term
commitment to furthering our understanding of the role of beliefs and attitudes
in human affairs, l have at the same time, perhaps more persistently than others,
deplored the fact that social psychologists continue by and large not to
appreciate the importance of values in human affairs" (p. 261 ). Rokeach devoted
considerable space to the concepts of attitude and belief and their roles in the
shaping of behavior. but he stated, "In my view, the attitude-behavior relation
question is a narrow one that can be subsumed under the broader question of
the relation among values, attitudes and behavior" (p. 271).

Social scientists are not in agreement regarding a single component of
human nature that might serve to predict subsequent behavior. The debate
regarding free will and determinism, as noted earlier, will not end soon.
However, consistent with the theoretical background provided in Chapter I, and
a review of the literature regarding culture, values, and the relationship between
values and behavior, the idea has been supported that values, and more

particularly value patterns, are one possible explanation for the variance in

62

school-absence behavior. Value patterns guide the behavior of parents, who, in

turn, guide (or in the case of small children, control) the actions of their offspring.

IbeAssessmembfllalues

Given that there is sufficient support forthe investigation of parental value
systems relating to the school-absence behavior of young children, a problem
arises concerning the measurement of those values. Underlying this
measurement is the definition of values, for there can be no precise appraisal
without such a definition.

Rokeach’s (1973) definition of values relied on the concept of "enduring
beliefs" regarding desirable end-states or modes of behavior. Thus, values are
in fact beliefs, but they are a special variety of beliefs. Values are beliefs that
have a greater degree of longevity, although they are changeable for an
individual, and they have a greater degree of generalizability, in that they account
for or "cover" a greater number of social behaviors than lower-level, less-
enduring beliefs.

The preceding definition is one of several in the field. The Allport-Vernon-
Lindzey (1960) instrument, though purported to be a values survey, was
designed to assess "the dominant interests of the personality" (Rokeach, 1973,
p. 36). The difference in levels of abstraction between this and the Rokeach
concept is revealed in the Allport-Vernon-Lindzey instrument asking the
respondent to tell "what I like," whereas the Rokeach survey involves judgments

of relative importance. Allport-Vernon-Lindzey "values" are captured in the

63

Rokeach definition of attitudes, particularly as the Allport-Vernon-Lindzey
"values" relate to comparatively specific social behaviors.

Kluckhohn and Strodtbeck (1961) advanced a definition of values that
relied on opinions related to the fundamental nature of reality. Their five
dimensions (good or evil human nature; subjugation to, harmony with, or mastery
over nature; past, present, orfuture time perspective; being, being-in-becoming,
or doing; and linearity, collaterality, or individualism) are, in Rokeach’s (1973)
words, "somewhat far removed from what we ordinarily mean by a ’conception
of the desirable’" (p. 22). In other words, the Kluckhohn and Strodtbeck concept
of values is perhaps one level of abstraction higherthan Rokeach’s. In 1973, he
wrote that the Kluckhohn and Strodtbeck dimensions "can be more aptly
described as basic philosophical orientations than as value orientations" (p. 22).

The definition of values used in the current study, including the notion of
being somewhere between beliefs and philosophical orientations on a continuum
of abstraction, is Rokeach’s. Rokeach’s definition was used because no other
conceptualization or instrument combines as concise a format with as

comprehensive an approach.

IbeBokeaehAlalueSumey

The 194 citations in Euros (1978) regarding the Rokeach Value Survey
include 116 journal articles and 78 theses. Thesis topics included the
relationships among value systems, attitudes, and interpersonal attraction

(Beech, 1967); the developmental structure of moral judgment (McLellan, 1970);

64

value systems and the priesthood (Hague, 1968); value systems and behavioral
change following a public executive workshop (Henderson, 1973); and value
patterns as they relate to social inheritance and educational systems (Robinson,
1977). The journal articles covered a wide range of topics, such as values and
job motivation (Brown, 1976), self-actualization and value structure (Mahoney,
1974), and values and neurosis (Mahoney, 1977). However, there was no
treatment of the topic at hand—the relationship of parents’ value systems to the
school-attendance behavior of young children.

A body of literature exists concerning the Rokeach Value Survey (RVS)
itself. Cohen (cited in Buros, 1978) described the ipsative nature of the
instrument and argued that this flaws the measurement accuracy because an
observer cannot discern the absolute strength of conviction expressed by two
people responding to the instrument. The first person’s fourth rank may equate
with the other person’s fourth, eighth, or fifteenth rank; one cannot tell. However,
Cohen (cited in Buros, 1978) later wrote, "Many research studies are reported
and the infirmity of the ipsative measures is successfully overcome by the
typically large sample sizes to produce statistically significant results" (p. 1032).
Cohen said that the RVS could be an effective research instrument, but he
warned against its use "in individual assessment in counseling, psychotherapy,
and selection" (p. 1032).

Kitwood (cited in Buros, 1978) also was concerned with the ipsative

structure of the RVS, again because the instrument cannot be sensitive to

65

differences in the intensity with which respondents hold values. Of course, the
assumption that Likert-type scales avoid this problem altogether is true only if
one person’s "Strongly Agree" is as strong as another person’s.

Kitwood also mentioned that the test-retest reliability of the RVS is low, but
not unduly so. He acknowledged that the items are not easily reducible to a
smaller number because of small correlation coefficients. He argued that the
strict rank ordering of each value relative to the others lends an artificiality to the
scale, citing the lack of conflict between, say, inner harmony and wisdom.
Kitwood also believed that there are some problems with the RVS items being
somewhat general in nature, and thus open to a variety of interpretations by
respondents with dissimilar backgrounds.

Mueller (1974) supported this last argument with his work on equalityand
freedom. He fleshed out definitions for the two words, making sure his college-
attending respondents clearly understood both terms. Mueller’s own equality and
freedom scales then intercorrelated .43 (p < .01), whereas Rokeach reported an
intercorrelation half that size. This result can only be interpreted to mean that,
with the additional definition Mueller gave to the constructs, the respondents
viewed them less distinctly. This does not help his argument that Rokeach’s
scales are invalid because of the fuzzy definition of the terms.

Kitwood (cited in Buros, 1978) also mentioned that there are "some
curious omissions" (p. 1033) from the RVS, yet he mentioned only one--"truth"

is not on the list. However, Kitwood went on to say:

66

Despite these weaknesses, the Rokeach Value Survey is more directly

concerned with values, as philosophically understood, than most, if not all,

other available instruments. It can at least be recommended as a general
probe into values for use with respondents whose academic attainment

is average or above. (p. 1033)

The concern that the RVS not be used with respondents whose academic
attainment is below average is countered by its frequent successful application
in studies involving just such populations. Feather and Cross (1975) studied the
generation gap between the value patterns of delinquent and nondelinquent boys
and their parents. McCarthy (1972) also studied the generation gap, but among
Catholic high school seniors and their parents. Toler (1974) analyzed the value
patterns of alcoholics aged 22 to 49 averaging 11.5 years of schooling, and of
addicts aged 21 to 35 averaging 12.4 years of schooling. Most participants were
Vietnam veterans. Carroll (1973) administered the RVS to high school students,
the public, and school officials in Appalachia. Jenkins (1974) used the RVS to
study the differences between drug abusers and nonabusers in the seventh,
eighth, and ninth grades in Arizona. Thus, there is ample precedent for using the
RVS in the current study with a population of mothers who had not completed
high school.

Kitwood and Smithers (1975) collaborated on an appraisal ofthe Rokeach
approach. In addition to the criticisms contained in Buros’s (1978) work, they
took issue with Rokeach’s assertion that the ranking task fell within Miller’s

(1956) suggestion that the human mind can process seven categories

simultaneously, give ortake two. Kitwood and Smithers mentioned that the scale

67
is, in fact, 18 items long, and they argued that about half of the job (the first half,
when the choices are many) is beyond people’s powers and half within.
However, there have been no reports of difficulty from thousands of respondents,
many with limited schooling.

Kitwood and Smithers also argued that there is "yet little evidence about
the way values are related to . . . human action" (p. 178). They believed that
Rokeach was operating on too many assumptions and cited their own work
(Kitwood, 1974) as an indication that "behavior is rarely a direct expression of
values" (p. 178). They pointed to new approaches (notably their own) and
cautioned those who study values that an instrument may not reveal what it
purports to show, although they were not explicit in accusing the RVS of this.
Counter to their argument is the theoretical framework of this study as described
in the first chapter, which supports a strong relationship between values and
behavior.

Greenstein and Bennett (1974) studied order effects in Rokeach’s survey.
Their concern was that Rokeach examined the problem only indirectly, and that
he used only one ordering (alphabetical) of the six thousand trillion (18 factorial)
possible orderings. Greenstein and Bennett therefore generated 218 different
random orderings of the RVS for 218 respondents. They then correlated the
vector of orderings for each respondent’s questionnaire with the vector of
rankings each person assigned to the items. Their results were unambiguous:

The fact that the observed mean correlations are so small allows us to
conclude that, for all practical purposes, Rokeach’s Value Survey is free

68

of order effects. In fact, only about .1% of the variation in responses may

be directly attributable to such bias. This amount of bias due to

presentation order seems acceptable. (p. 395)

Mahoney and Katz (1976) conducted an investigation to "explore the
structural bias of the survey, countenancing the ipsative nature of the data" (p.
205). They hypothesized that the analysis would yield a relatively small number
of factors that would relate closely to what they termed "established differences
in ideological orientations" (p. 205). The researchers administered the RVS to
120 volunteers from introductory psychology classes in a mass-testing
procedure. They then calculated Spearman rank-order correlations for each pair
of values in the 36—by-36 matrix. Factor extraction was performed on the
correlation matrix, and the factors were then subjected to varimax rotation. This
procedure yielded 13 factors. Mahoney and Katz concluded that the RVS could
be reduced to a smaller number of "underlying structural factors" (p. 210). The
problems with this conclusion are how to label these factors, how then to
measure them, but most important whether they in fact exist.

For example, the first factor identified through this process accounted for
19.2% of the variance. The positive pole was defined by imaginative, intellectual,
independent, and inner harmony. The negative pole consisted of clean, polite,
and a comfortable life. Mahoney and Katz argued that this factor reflects an
"education versus economic" orientation, and they acknowledged that Rokeach
thought so too. But what to name this factor? And how, besides the Rokeach

method, is one to measure it?

69

A second example is their second factor (15.7%), with an exciting life,
imaginative, and courageous on the positive side and negative loadings on
forgiving, salvation, and helpfirl. Again, Rokeach (1973) mentioned this factor as
"personal competence versus religious involvement." Although these two
factors accounted for more than one-third ofthe variance using the Mahoney and
Katz method, the authors put forth no suggestion as to how either factor might
be measured more effectively.

There is a more fundamental problem with the Mahoney and Katz logic.
Although it is admirable for the points made from a strictly measurement
perspective, the authors missed an important finding in their own data.
Imaginative appears on both Mahoney and Katz factors 1 and 2, yet the
researchers offered no explanation for how one value can represent, in part, both
"personal competence" and "education." Is there some relationship between
"education" and "personal competence" that Mahoney and Katz missed? The
arguments set forth in the first chapter of the current study regarding the five-
value model suggest that indeed there is. Mahoney and Katz’s attempt to
separate their factors 1 and 2 on the basis of factor analysis erred on the basis
of a correlation between personal competence and education that people
express when they rank imaginative on the RVS. If in fact imaginative is a
"member" of both the "education" and "personal competence" camps, as
Mahoney and Katz’s data suggest, their factors 1 and 2 are not separate. Yet

neither their data nor their conclusions suggest a "super factor" linking the two.

70

The argument that the RVS is reducible to a smaller number of factors is
laudable but untenable given this conflict.

The most serious criticisms ofthe Rokeach instrument center on ipsativity
and whether a different number of values, or other values, exist. Regarding the
former, Rokeach argued that uncovering a pattern or hierarchy necessarily
involves choosing one over another. But the survey is as comprehensive and as
manageable as possible, given the complexity of the constructs under study.
Thus, the Rokeach Value Survey stands as a reasonable instrument to use in

measuring value patterns.

Summary

In the literature reviewed in this chapter, absenteeism from school was
seen as a problem that exists across time and cultures; this problem is more
prevalent among poor, urban populations than more affluent suburban groups.
Some investigators have posited that the causes for absenteeism are lodged in
parents’ orientation toward the value of schooling and the importance of school
attendance. The effects of high rates of absenteeism were reviewed as being
essentially negative, including short-term academic deficits and long-term
problems such as continued school absence, delinquency, and reduced adult
income.

Values were described in the literature as being controlling in the
determination of behavior. Research regarding a subculture (the poor)

interfacing with the mainstream culture was discussed as a model forthe role of

71

cultural conflict, the context of the current hypotheses. Literature detailing the
notion that values can be defined, and therefore assessed, was reviewed.
Finally, studies regarding the Rokeach Value Survey were discussed; the

instrument was considered to be useful for its intended application in this study.

CHAPTER III
METHODOLOGY

Introduction

The researcher’s purpose in this study was to investigate whether there
is a relationship between parental values and children’s school absenteeism.
The theory is that it is possible to predict or explain the school-attendance
behavior of early elementary school children given the value patterns held by
their parents. The methodology used in conducting the study is explained in this
chapter. The setting in which the study was carried out is described first,
followed by a description of the population and the sample-selection process.
The distinction made between excused and unexcused absences also is
discussed. The survey techniques are explained next, followed by a description

of the data-analysis methods.

Ibefiettingibntbefitudy
The school district in which this study was conducted is located in the
center ofa metropolitan area with a population of more than 250,000, a little over
half of whom reside in the city. The major employers in the area are an

automobile manufacturer, the state government, and a major university. The

72

73
school district includes 33 elementary schools, 4 middle schools, and 3 high

schools; it employs 1,200 teachers and serves about 20,000 students.

Eilm_Stud¥

The selection of the Rokeach Value Survey as the instrument to assess
the values of parents in this study resulted in the need to pilot the survey with a
group of mothers who had not finished high school. The potential problem was
that the reading and comprehension requirements of the instrument might be too
high for nonfinishers of high school.

The researcher had a group of five mothers, none of whom had high
school diplomas, who worked as volunteers in an elementary school. The RVS
was administered to them with no verbal directions. Following their completion
of the instrument, a debriefing conversation was used to probe any difficulty the
subjects might have encountered.

The problems expressed by the subjects related to the difficulty of
choosing between two values thought to be close In ranking. There were no
comments regarding any of the terms, and no other comments regarding the
difficulty of the task. The pilot allowed the researcher to proceed with greater

confidence regarding the appropriateness of the instrument.

74
IneElelatibnandjample

The population from which the sample was drawn comprised all
kindergarten through third-grade students whose mothers had not completed
high school, a total of 1,604 cases.

In selecting the sample, the researcher addressed several problems that
were central to the questions of interest. There was the need to identify a
sufficiently large number of persistent absentees, so that enough "no excuse"
subjects could be found. And it was clear from the theoretical background that
the youngest elementary school students would be the focus of the investigation
because the researcher wanted to be as certain as possible that parental
influence was paramount in controlling school absence. In addition, some
mechanism for controlling for income differences had to be determined because
Rokeach (1973) found that such differences were strongly associated with
variations in value patterns.

Regarding the issue of how many persistent absentees were likely to be
found, and of those the number whose absences would be considered
unexcused, Hansard (1974) found that 2.2% of the secondary students studied
had no legitimate reason for absence; the Scottish Education Department (1977)
found 1.6% of male absences and 1.1% of female absences to have been
without excuse; and Galloway (1980) reported a persistent-absence group of
elementary students to be about 0.35% of the total group (N = 10,858). These

proportions were expected to be somewhat higher in urban areas and among the

75
poor (see Chapter II), so a figure of 5% was used in planning for this study,
based in part on the literature and in part on the writer’s observations relative to
the prevalence of high-absence patterns in the school district involved in the
study.

Cohen (cited in Buros, 1978) suggested that "the infirmity of the ipsative
measures [of the RVS] is successfully overcome by the typically large sample
sizes to produce statistically significant results" (p. 1032). He mentioned the
multitude of studies in which the Value Survey has been used. In those studies
(cited in Chapter II), and including Rokeach’s national sample, no subgroup had
fewer than 30 members. This number was selected as the minimum target for
each subgroup in the current study.

Given a minimum of 30 high-absence, poor-excuse students’ mothers
participating in the study, and using the estimate of 5% asthe proportion of these
in a poor, urban student body, a population of at least 600 mothers would need
to be identified (600 x .05 = 30). But low-absence students and medium-
absence students were needed as well, so the figure was doubled to 1,200 to
ensure a large enough population from which to secure the sample.

To control for value-pattern differences that are known to occur due to
income differences (Rokeach, 1973, p. 144), the study population was drawn
from mothers with similar incomes. Coupled with the indication that high
absenteeism is more prevalent among the poor, mothers with the lowest incomes

were selected for the study. To select mothers with the lowest incomes, and

76

because income information was not directly available, a proxy was used.
Mothers who reported not having attained a high school diploma formed the

population for the study.

AbsenceBate

Some method had to be devised to ensure a range of absence rates
among the students included in the study. Thus, the concept of low-, medium-,
and high-absence categories was employed. Sampling from thethree categories
was necessary to ensure proportionate representation and a range as desired.

After Hansard (1974), the Scottish Education Department (1977),
Galloway (1980), the Educational Research Service (1977), the Children's
Defense Fund (1974), Berg, Brown, and Hullin (1988), Moore (1966), Newson
and Newson (1977), and the Central Advisory Council for Education (1967), the
researcher established definitions of low, medium, and high rates of attendance.
These rates were somewhat arbitrary, based on the wide variances reported in
the literature (see Chapter II). However, combining the logic in the literature with
the researcher’s field experience supported the definitions used in this study.

In conversations with the writer, teachers frequently have claimed that
students who miss a day of school each week are more difficult to teach than
students who are usually in school. Often a particular student will be gone every
Monday, or every Friday, or will arrive at school mid-morning or later twice or

three times a week, and thus will miss a significant amount of important

77

instructional time. This suggests that missing school one day out of five, or an
absence rate of 20% or so, would be an appropriate limit to define the high-
absence student. This rate is also generally consistent with the definitions given
by the authors cited above. However, the researcher decided on 16% for the
high-absence rate, to ensure that all students who missed a lot of school would
be included in the study. From a 180-day school year, this translates to 30 days
of absence, which became the line separating high absence from medium
absence.

The line differentiating low absence from medium absence was drawn at
the 5% level. Thus, students missing from zero to nine days of school during a
180-day school year were judged to have a low absence rate. Students missing
from 9.5 to 29.5 days had a medium absence rate. Students with 30 or more
days of absence were members of the high-absence group. Viewed another
way, students who missed school less than a day per month were low absence,
those who missed more than a day a month to almost a day every week were
medium absence, and those who missed close to a day every week or more

constituted the high-absence group. These categories are detailed in Table 3.1.

78

Table 3.1: Definitions of absence categories, based on 180 days of possible
auendance.

 

Low-Absence Medium-Absence High-Absence
Days 0-9.0 9.5-29.5 Over 29.5
Percent 0-5% 5.3%-16.4% Over 16.4%

 

 

 

 

 

 

 

 

Absence information was collected atthe elementary school buildings and
entered into a central computer. At the time of this study, the school district was
in the process of having secretaries change from pencil-and-paper reporting
(encoding "bubble sheets") to entering the information directly onto a computer.
All data regarding absences were then available through the central database.

An initial computer search for students in grades 1, 2, 3, and 4
(kindergarten was added and grade 4 discarded later) whose mothers reported
no high school diploma yielded 1,529 cases, which compared favorably with the
target of 1,200 cases. Of these, 812 were low-absence, 619 medium-absence,
and 98 high-absence students. The proportion of high-absence students was

slightly greater than predicted (see Table 3.2).

Table 3.2: Distribution of the population by absence category: Grades 1
through 4 (N = 1,529)

 

Low-Absence Medium-Absence High-Absence

 

Number of cases 812 619 98

 

 

 

 

 

 

Percent of total 53.1% 40.5% 6.4%

 

 

 

79

Kindergartners were not included in the initial search because the
absence coding for this grade was different from the others. The discovery that
a correction could be used to equate kindergarten absence reporting with the
other grades prompted a second search, which resulted in the inclusion of
kindergartners and the exclusion of fourth graders.

The use of the youngest students possible as the study population was
critical. This study was concerned with students who were allowed to be away
from school with their parents’ consent (detailed as the concept of "absenteeism"
in Chapter I), whereas the notion of "truancy" involves students staying away
from school on their own. The researcher’s experience with elementary school
students has shown that truancy is known at the fifth-grade level, rather
uncommon at the fourth-grade level, and rare among third graders. Thus, by
excluding fourth graders and including kindergartners in the population, greater
assurance was provided that school absence is absenteeism controlled by the
parent, and not truancy controlled by the student.

Given the results of the initial computer search of grades 1 through 4, the
writer was confident that changing to a K-3 population would still provide an
adequate population for the study. The kindergarten group with mothers
reporting no high school diploma (n = 420) was thus included in the study, and
the corresponding group of fourth graders (n = 345) was excluded. The sample
could then be drawn from the new population of1,604 cases. It is interesting that

the high-absence category gained 50 cases when the 420 kindergartners were

80
included in the study. The proportion of kindergartners in this high-absence
group, 11.9%, was nearly double the proportion of older students who were
highly absent, which was 6.4%. The kindergarten through grade 3 population

from which the sample was finally drawn is shown in Table 3.3.

Table 3.3: Distribution of the population by absence category: Grades K
through 3 (N = 1,604)

 

Low-Absence Medium-Absence High-Absence
Number of cases 775 701 128
Percent of total 48.3% 43.7% 7.9%

 

 

 

 

 

 

 

 

To secure a sample of at least 30 cases in each of the three absence
categories, systematic sampling (Hopkins 8 Glass, 1978, p. 187) was employed
in the low- and medium-absence groups. All cases in the high-absence group
were included in the initial sample (where absence was unexcused—see
discussion below).

The systematic sampling technique involves selecting a random number
between 1 and 10 from a table of random numbers, picking the case that
corresponds to that number as the starting point, and then selecting every nth
case to arrive at the desired number of cases. Because 30 respondent cases
were needed in the low-absence category, and because response rates are

generally 70% in studies involving questionnaires (Babble, 1973), and

 

81

anticipating that the response rate for the target population in this study might
well be lower than usual, the researcher decided to select every ninth case from
the close to 800 cases in the low-absence category. It was thought that this
process would yield close to 100 cases for the sample.

The medium-absence category consisted of two groups of students—those
whose absences were excused, forthe most part, and students whose absences
were not generally excused. Of primary interest in this study was the group of
students whose absences were not generally excused, forthe reasons discussed
below. Thus, faced with a population of some 700 students in this medium-
absence category, the researcher selected every fifth case, generating more than
150 cases, some of which subsequently were eliminated on the basis of reason
for absence.

All cases in the high-absence group were included in the preliminary
sample; that is, no systematic sampling was employed. Instead, reasons for
absence were determined, and those students whose absences were primarily
unexcused were included in the final sample. Also, some sibling duplications
and blended-family situations (students with different names living in the same
household) further reduced the number of cases in all absence categories before
the Value Surveys were mailed.

Names of the 1,604 population cases were arranged on the computer
printout by school, then on the basis of absence category, then in alphabetical

order by the last name of the student. When systematic sampling was employed.

82

a school with only six low-absence students whose mothers reported no diploma
may have contributed no students to the sample, or perhaps one, depending on
where that school was "positioned" along the continuum of absence categories.

The population represented 31 of the 33 elementary schools in the urban
district where the study took place. Two schools were excluded from the study
to ensure complete separation between the researcher and the respondents.
The researcher was the principal at one building and was to be transferred to the
second building for the next school year.

The final sample to whom surveys were mailed numbered 226 cases.
When the techniques described above were used, the low-absence group
included 84 cases, the medium-absence group had 64 cases, and the high-

absence group included 78 cases (see Table 3.4).

Table 3.4: Distribution of the sample by absence category (N = 226)

 

Low-Absence Medium-Absence High-Absence

 

Number of cases 84 64 78

 

 

 

 

Percent of total 37.1% 28.3% 34.5%

 

 

 

ExcusednlensuaUnexcusedAbsences
The question of whether a student's absences are essentially excused or

not seems easily answered. All one needs is a doctor’s note or a note from the

83

parent detailing the reason for absence. And, in fact, among low-absence and
many medium-absence cases, this procedure is common. However, among
some medium-absence students and many high-absence students, such is not
the case.

Because many families of high-absence students and some families of
medium-absence students rarely support their children’s absences with
documentation, knowledge of the true reason for absence typically is not known.
Thus, one must rely on a more intimate knowledge of the family and child than
is commonly supposed. This is why many authors (Bolton, 1977; Farrington,
1980; Fogelman, 1976; ISTD, 1974) have recognized the teacher as a reliable
reporter of the "real" reasons behind absence patterns. The relationship
between the teacher and the student, and at least at the elementary school level
between the teacher and the parent, is intimate enough to allow the teacher to
determine with some certainty why a student is absent.

In the district in which this study was conducted, school secretaries
routinely call parents when students are absent. Some years ago, several
incidents of temporarily "lost" elementary students, along with changes in
secondary student attendance-tracking practices, resulted in clear directives to
the elementary schools to contact families when children were absent.
Subsequently, secretaries acquired knowledge of family circumstances as they
communicated with both the family and the teachers regarding student

attendance. In addition, many school secretaries in this district work closely with

84

lunchroom cashiers to establish accurate counts for the number of hot lunches
to be ordered from central suppliers. This task must be done daily, and the
counts must be accurate. Therefore, nearly every secretary knows on a given
day who is in school, who is not, and who is likely to show up just before lunch.
Accordingly, the researcher spoke with secretaries in every building in
which medium- and/or high-absence students had been identified forthe sample.
All of the secretaries knew the writer, so introductions were brief and initial
rapport was high. Each secretary was told that the call concerned a study the
writer was conducting regarding students who missed a lot of school. Many
secretaries immediately expressed interest and appreciation that someone was
looking into the problem. Each secretary was told that the writer would read her
a list of names and, if she was willing, would ask her to indicate whether the
students’ absences were mostly excused or unexcused. Every secretary was
willing to proceed. The writer then read the following text to each secretary to
ensure that all of them were given the same directions:
Some students are gone a lot, but you know there’s a good reason.
Chicken pox or a long flu bout, a cold and cough that won't go awaynyou
know the illness things. Also, some kids may have gone on a trip with
their parents--a vacation--but you know the parents got work from the
teacher and did "school stuff" with the kids when they were gone. So
these kids are excusednout of school, but with good reason. There are
others l’ll name, however, who are out a lot but really don’t have excuses.
They say they’re sick, but you know they’re not. They miss a half day
because "it was just too hard to get awake." They go on trips with Mom,
but it’s no vacation. Mom had to get away from the boyfriend who was
threatening her, or she’s temporarily with her girlfriend because the rent

is due. These kids should be in school, but they’re not. Their absences
are primarily unexcused. All you need to do, when I read the names, is

85

to say either "excused" if the student is mostly excused, or "unexcused"
if that’s the case.

After the first few calls, it was obvious that the secretaries had the working
knowledge to support their judgments regarding the reasons for students'
absences. In case after case, they knew not only the name of the student, but
they also could guess the number of days absent, the given reason, and the real
reason; also, the secretaries remembered such details as difficulty reaching the
home, the absence patterns of siblings, the teachers’ reactions, and the effect on
the students’ school progress. In fact, the secretaries were so enthusiastic in
sharing their knowledge about the cases that the writer added the following
sentences to the directions:

I selected only a few students from each school, so perhaps I don’t have

your toughest cases. When I finish with my list, feel free to tell me about

who I missed.
Any additional students the secretaries mentioned were not added to the sample,
but the extra discussion was allowed in order to extend the initial rapport
between the writer and the secretaries. In fact, additional time had to be
budgeted into this phase of the research due to the interest and volubility of the
secretaries on the topic. Because of this turn of events regarding the

determination of reasons for absence, the resultant sorting of students into the

"unexcused" category proceeded with greater certainty than anticipated.

86
SunLevlecbnibues

On the Rokeach Value Survey, each respondent is asked to arrange 18
terminal values in order of importance to the individual (see Appendix B). The
terminal values are arranged alphabetically on 18 gummed labels, which can be
peeled off the right column and stuck to the left column. After placing the labels
in the preferred order on the left column, the respondent may change the order
simply by removing and reapplying the labels. The labels are not permanent.
The respondent then turns the page and repeats the process with 18
instrumental values.

A total of 226 surveys were mailed to the parents selected for study; in
most cases, the surveys were addressed to the mother. In a few instances the
father was listed as "Parent 1" on the student enrollment card as the parent to
call first in an emergency, the data from which the mailing list was culled. The
intention was to connect with the parent having the most day-to-day contact with
the student so as to maximize the effect of the parents’ values on the student.
A letter explaining the study (Appendix C) was included with the survey. The
letter included an invitation to the parent having the most interaction with the
student to complete the survey. Also included in the mailing was a retum-
addressed, stamped envelope in which respondents were asked to return the
completed instrument.

The surveys were mailed in late November 1992. By mid-December, 14

surveys had been returned. Ten other surveys were returned by the post office,

87
stamped "undeliverable" or "addressee unknown." The addresses were double-
checked in the computer system, and three of the surveys were remailed. Seven
families had left the area, and no forwarding address was available for them.

For the next seven months, the researcher made a series of telephone
calls and home visits to secure additional returns. The writer hired and trained
an assistant, who followed up on nonrespondents in the same manner. The two
main problems encountered during this collection phase were the number of
residence and telephone changes for the families, as well as respondents who
agreed to complete and mail the survey but never did so.

Among thethree absence-rate groups there were 108 residence changes,
including 21 families who moved twice and one family who moved three times
over the nine months of the study. There were 102 telephone problems,
including new acquisition (46), no telephone or wrong number reported to the
school (21 ), and phones being disconnected (35). Thus, simply locating a family
to encourage return of the survey was difficult. This problem was not
unanticipated, however, which was why a general follow-up mailing was not
conducted. Rather, subsequent mailings were made only to those families
contacted by telephone, and home visits were made to those not spoken with by
telephone.

An interesting facet regarding the number of moves and telephone
problems was the disproportionate number of such cases within the high-

absence group. More than half of that group moved, whereas just over a third

88
of the medium-absence group did, and fewer than a fourth of the families in the
low-absence group changed residence in nine months. Overall, 42% of the
sample of 226 families moved during the study period, and 45% either had no

phone or had a phone change during that time (see Table 3.5).

 

 

 

 

 

Table 3.5: Residence changes and telephone problems within the three
absence groups.
Moved New Phone No Phone Phone
Disconnected
Low absence 25 12 2 8
(n = 84) 30% 14% 2% 10%
Medium absence 27 13 4 12
(n = 64) 42% 20% 6% 19%
High absence 43 22 15 15
(n = 78) 55% 28% 19% 19%
Totals 95 47 21 35
(N = 226) 42% 21% 9% 15%

 

 

 

 

 

 

 

Note: Figures do not add to group size or 100% because any family could be in
no cell, any cell, or all cells appropriate to the group.

The task of collecting surveys from this population became one of
checking the latest school district computer update of nonrespondent parents.
Then, if a phone was listed, a call (and in some instances several calls) was
placed to the parent. The writer or assistant reminded the parent about the
mailing and asked if they could provide any help so that the finished survey

would be mailed. In many cases, parents remembered the survey but had

89

thrown it out, so a new one was either mailed or delivered to them, depending on
their preference. In some cases, respondents remembered the survey and
promised to complete it. A date was mutually established in these cases, and if
the researcher did not receive a completed survey, another contact was made.
Still other respondents did not remember the survey, so another was either
mailed or delivered to them.

If the computer check showed no telephone, the "Emergency Contact"
data screen was accessed for the family. If a number could be found at which
a message could be left for the parent, this was done. If not, the writer or
assistant visited the family’s home and tried to have the parent complete the
survey at the time of the visit. In several cases, the family did not live at the
given address, and neighbors were asked about their location, but generally to
no avail. In these cases, the school secretary (at the school where the child still
attended) was contacted; in every case the secretary suspected that the family
had moved but had no way of confirming it. No further attempt was made to get
these parents to complete the survey.

As a result of these efforts over a nine-month period, 90 surveys were
collected--32 from the low-absence group, 30 from the medium-absence group,
and 28 from the high-absence group. Efforts to secure additional surveys were
suspended at that point because there were enough returns to proceed with the
study. The numbers and percentages of surveys returned by respondents in the

three absence groups are shown in Table 3.6. As can be seen in the table, the

90

low- and high-absence groups had return rates exceeding one in three, whereas
the medium-absence group returned nearly half of the surveys. The overall

return rate, 90 surveys from the 226 that initially were mailed, was 39.8%.

Table 3.6: Surveys sent and returned (N = 226).

 

 

 

Low- Medium- High- Totals
Absence Absence Absence
Number sent 84 64 78 226
Number returned 32 30 28 90

 

 

 

 

 

 

 

Return percentages 38.1% 46.9% 35.9% 39.8%

 

One question of interest was the ethnicity of the sample as compared with
the larger group of families where there was a report that the mother had not
finished high school. The ethnicity of parents was not identified, but the
information parents supplied to the school regarding the child’s ethnicity was
available. Tables 3.7, 3.8, and 3.9 show the ethnic composition of the

population, the sample, and the returns, respectively, through the use of data

regarding the ethnicity of the child.

91

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 3.7: Ethnicity of the population (N = 1.604).
Ethnic Group Low-Absence Medium- High-Absence Totals (%)
Absence
Native Amer. 12 12 0 24 (1.5%)
Black 160 205 37 402 (25.1%)
Asian 138 28 3 169 (10.5%)
Hispanic 134 169 39 342 (21 .3%)
Caucasian 329 287 51 667 (41.6%)
Totals (%) 773 (48.2%) 701 (43.7%) 130 (8.1%) 1,604 (100%)
Table 3.8: Ethnicity of the sample (N = 226).
Ethnic Group Low-Absence Medium- High-Absence Totals (%)
Absence
Native Amer. 0 1 0 1 (0.4%)
Black 20 25 27 72 (31.9%)
Asian 11 1 1 13 (5.8%)
Hispanic 12 7 22 41 (18.1%)
Caucasian 41 30 28 99 (43.8%)
Totals (%) 84 (37.2%) 64 (28.5%) 78 (34.5%) 226 (100%)

 

 

 

 

 

 

 

 

92

 

 

 

 

 

 

 

Table 3.9: Ethnicity of the returns (N = 90).
Ethnic Group Low-Absence Medium- High-Absence Totals (%)
Absence

Native Amer. 0 0 0 0 (0.0%)
Black 7 11 13 31 (34.4%)
Asian 0 0 0 0 (0.0%)
Hispanic 9 4 6 19 (21.1%)
Caucasian 16 15 9 40 (44.4%)
Totals 32 (35.6%) 30 (33.3%) 28 (31.1%) 90 (100%)

 

 

 

 

 

 

 

Interpretation ofTable 3.10, which shows the combined ethnicity analysis,
reveals that the ethnic composition of the group that returned surveys was not
unlike that of the original population from which the sample was drawn. Blacks,
Caucasians, and Hispanics were slightly more numerous among the returns than
among the population. Native Americans and Asians accounted for a combined
12% of the population, but only 6.2% of the sample. No surveys were returned
from these groups. With regard to the Asians in the sample, the investigator
made several home visits in order to secure survey returns. In each case, the
parents’ limited proficiency in English required extensive translation through their
elementary-aged children. The researcher concluded that the potential for bias
and/or miscommunication was too great, so no further attempt was made to

complete surveys from this group.

93

Table 3.10: Combined ethnicity analysis (in percent).

 

 

 

 

 

 

 

 

 

 

 

 

Ethnic Group Population (N = 1,640) Sample (N = 226) Returned (N = 90)
Native Amer. 1.5% 0.4% 0.0%
Black 25.1% 31.9% 34.4%
Asian 10.5% 5.8% 0.0%
Hispanic 21.3% 18.1% 21.1%
Caucasian 41.6% 43.8% 44.4%

D I -9 I . I I .

The first two research questions suggest descriptive analysis. Those

questions were:

1. How did parents in the current sample rank the values on the
survey?
2. How do these results compare to those from other samples?

Respondents ranked the terminal values from 1 through 18, with the rank
of 1 being the most important value. The process was repeated for the
instrumental values. A mean ranking for all parents in the study (N = 90) for each
value on the survey was then calculated. This was performed separately for the
terminal and instrumental lists. An ordering of the values from most to least
important could then be accomplished, given the mean ranking for each value.
Two lists (one for terminal and one for instrumental values) of "composite"

rankings were then available for examination.

94

These overall rankings were then compared to the composite rankings
reported by Rokeach (1973). The purpose of the comparison was to determine
whether the current sample resembled Rokeach’s national sample and to note
any differences. The comparison employed Rokeach’s non-high-school-
graduate respondents because that was the educational attainment of the current
sample.

Regarding the hypotheses, the analyses performed relied on
nonparametric statistics. The Rokeach Value Survey yields ordinal data because
respondents rank the various values included in the survey, so nonparametric
statistics were appropriate forthese data (Hopkins 8 Glass, 1978; Siegel, 1956).

The test forthe first hypothesis, which addressed the possible relationship
between rankings of values and student days of absence, was accomplished
through the use of the Spearman rank correlation coefficient, or rho. It requires
a calculation involving the sum of squares of differences between the ranking of
a value by parents and the rank of a student’s absences. Thus, all parental
rankings of each of the values were analyzed against the ranking of each
student’s days of absence. It should be noted that information within the
absence data was lost in the conversion to ranks. The absence data are of
interval scale; thus parametric statistics could have been used. However, the
other variable in the comparison, value rankings, was only of ordinal level, so the

nonparametric statistic was appropriate (Siegel, 1956).

95

The test used for the second hypothesis, concerning the possible variation
in the way parents ranked the values on the survey, was the Kruskal-Wallis one-
way ANOVA. All respondents’ rankings (from 1 through 18, inclusive) for a given
value were arranged from the highest to the lowest ranking. Each was then
reassigned a new rank corresponding to its position among all 90 respondents.
The computer then calculated group means for the three absence groups (low-,
medium-, and high-absence), which were then analyzed to determine whether
any differences existed.

The Kruskal-Wallis test revealed whether differences existed among the
three groups, but it did not provide sufficient precision to ascertain whether
differences between any two group means were significant. The Mann-Whitney
Utwo-sample test is an appropriate test to use "if the data are observed as ranks
and not as test scores or some other measured variable" (Marascuilo 8
McSweeney, 1977, p. 6). Thus, the Mann-Whitney was used as a follow-up test
to compare pairs of means where the Kruskal-Wallis revealed values were
ranked differently among the three groups of parents.

The third hypothesis focused on the 12-value model. Directional
predictions forthose values were made based on Rokeach’s results considering
years of schooling of the respondents. The accuracy of the 12-value model was
assessed given the combined results of the Spearman, KruskaI-Wallis, and
Mann-Whitney tests.

Results of all analyses are presented in Chapter IV.

CHAPTER IV

RESEARCH FINDINGS

Introduction

The researcher’s purpose in this study was to examine the possible
relationship between parental values and student absenteeism. Five research
questions were posed. The first two suggested descriptive analysis, and the final
three resulted in testable hypotheses. In this chapter, results of the
administration of the Rokeach Value Survey and subsequent inquiries are
presented. The chapter is organized into five sections. The first section provides
a preliminary analysis, which presents the general survey results and compares
these results to the Rokeach national sample of 1973. The second section
addresses the first hypothesis, and the next two sections treat the second and

third hypotheses, respectively. The findings are summarized in the final section.

l'l' IE' I'

In this study, 90 parents responded from families in which the mother did
not hold a high school diploma. Each parent ranked 18 terminal values in order
of importance to them from first, or most important, to eighteenth, or least
important. The process was repeated for the instrumental values. The arithmetic

mean rank was calculated for each value. Table 4.1 contains these results for

96

97

the terminal values, and Table 4.2 contains these results for the instrumental
values. Included in the tables are the highest rank received by each value, the
lowest rank received, the standard deviation, and the derived "composite rank."
This last item denotes the arrangement of the values according to the mean rank.
The lowest numerical mean rank (that value rated "most important") was
assigned the composite rank of 1, the next 2, and so on, including the highest
numerical mean rank (rated "least important") assigned the composite rank of 18.

The rankings forthe terminal values are presented in Table 4.1. The rank
of mean rankings for the terminal values was from a mean ranking of 6.49 for
health (the most highly ranked value to a mean ranking of 14.38 for national
security(the least highly ranked value). The mean ranks forthe values dropped
about four- or five-tenths of a point, progressing from most important to least
important. The exception was a drop of three points (from a mean rank of 11.36
to a mean rank of 14.38) between a world of beauty and national security.

Standard deviations ranged from a low of 3.44 (national security) to a high
of 6.10 (a world of beauty). Ten values were ranked with a standard deviation
between 4.16 and 4.93.

For 15 of the 18 values, the highest rank was first. For all values. the ‘
lowest rank was seventeenth or eighteenth. The highest ranking for wisdomwas
third, for a sense ofaccomp/ishment fourth, and for national securitythe highest

rank assigned by a parent was sixth.

98

Table 4.1: Means and composite rankings: Terminal values fortotal sample.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Value Mean Rank Std. Dev. Minimum Maximum Composite
Rank Rank Rank

A comfortable life 8.24 4.48 1 17 6
Equality 6.76 3.62 1 17 2
Exciting life 10.38 4.74 1 18 13
Family security 7.17 4.77 1 18 3
Freedom 7.98 4.38 1 17 5
Health 6.49 4.53 1 17 1
Inner harmony 11.03 4.64 1 18 15
Mature love 9.64 4.93 1 18 9
National security 14.38 3.44 6 18 18
Pleasure 10.10 4.90 1 18 11
Salvation 9.14 5.41 1 18 8
Self-respect 7.39 5.10 1 17 4
A sense of 11.09 4.29 4 18 16
accomplishment

Social recognition 10.83 5.31 1 18 14
True friendship 9.79 5.57 1 18 10
Wisdom 10.37 4.16 3 18 12
A world at peace 8.87 5.95 1 18 7
Aworld of beauty 11.36 6.10 1 18 17

 

 

 

 

 

 

 

 

99

The rankings for the instrumental values are presented in Table 4.2. The
range of mean rankings for the instrumental values ranged from 6.36 for clean,
the most highly ranked value, to a mean ranking of 13.29 for obedient, the least
highly ranked value. The differences in mean rankings between adjacent values
ranged from as little as one hundredth of a point (between the most highly ranked
clean, 6.36 and second-place finisher honest, 6.37) to nearly two points (fourteen
place self-controlled at 10.29 against fifteenth-place logicalat 12.28). Standard
deviations ranged from a low of 3.55 for imaginative to a high of 6.05 for polite.
Thirteen values were ranked with a standard deviation of between 3.55 and 4.84.

The highest ranking for 16 of the 18 values was first. The lowest ranking
for all values was eighteenth, except honest, which had a lowest ranking of
seventeenth.

The analysis must first take into account the educational levels of the
different samples. Rokeach’s national sample ranged in years of education from
zero years to post-college degree. The current sample ranged from zero years
to "some high school," meaning the respondent had not completed high school.
Table 4.3 shows the number and percentage of respondents in Rokeach’s
national sample who had not completed high school. Table 4.4 shows the same
information, with slightly different categories for years of education, for

respondents in the current study.

100

Table 4.2: Means and composite rankings: Instrumental values for total

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

sample.

Value Mean Rank Std. Dev. M:;::m M:;T:m Co::::ite
Ambitious 7.43 3.97 1 18 4
Broad-minded 9.84 4.82 1 18 12
Capable 8.86 4.53 1 18 8
Clean 6.36 4.77 1 18 1
Courageous 9.73 4.66 1 18 11
Forgiving 8.60 4.46 1 18 7
Helpful 8.42 5.20 1 18 5
Honest 6.37 4.74 1 17 2
Imaginative 13.23 3.55 5 18 17
Independent 9.42 5.31 1 18 10
Intellectual 12.62 3.56 1 18 16
Logical 12.28 3.57 2 18 15
Loving 6.57 4.34 1 18 3
Loyal 9.40 4.84 1 18 9
Obedient 13.29 4.79 1 18 18
Polite 9.87 6.05 1 18 13
Responsible 8.43 5.63 1 18 6
Self-controlled 10. 29 5.78 1 18 14

 

 

 

 

 

 

 

 

101

Table 4.3: Years of schooling (lowest of seven categories) of Rokeach’s
national sample.

 

 

 

 

 

 

 

Years of Schooling Number (%)
0-4 64 ( 4.6%)
5—8 263 (18.7%)
Some high school 320 (22.8%)

 

Table 4.4: Years of education of the current sample, by absence group (N =

 

 

 

 

 

 

90).
Years of Low-Absence Med-Absence High-Absence Totals (%)
Schooling
0 years 1 0 0 1 ( 1.1%)
1-6 years 2 1 2 5 ( 5.6%)
7-8 years 4 4 3 11 (12.2%)
Some h.s. 25 25 23 73 (81.1%)
Totals 32 30 28 90 (100%)

 

 

 

 

 

 

 

In both samples, the single largest category of respondents was "some
high school." The apparent spike in the percentage of "some high school"
respondents for the current study is explained by the controlled sampling, in
which only families in which the mother’s level of education was less than high
school completion were selected.

As shown in Table 4.4, the numbers of mothers reporting no high school

were similar for all absence groups. Combining the three levels of schooling

102
from zero to eight years inclusive, the low-absence group had seven members
and the medium- and high-absence groups had five members each at these low
levels of schooling.

To compare the Rokeach sample to the current sample, the composite
ranks were used. These ranks were generated from the median rankings
(Rokeach sample) and mean rankings (current sample). The rankings were
arranged in numerical order, with the highest-ranked value being assigned the
composite rank of one. All other values were assigned a composite rank in
order, ending with the lowest-ranked value, which was assigned the rank of 18.
Because Rokeach did not combine the three subgroups representing "no high
school diploma" and because the present researcher did, Table 4.5 shows the
figures for three of Rokeach’s groups in three columns and reserves a single
column for the current sample for the terminal values.

In alphabetical order (not in order of importance), the first difference of
note was the rankings for equality. Rokeach’s groups ranked this value seven
ranks lower than did the current sample. The current sample ranked health at
the top of the list (Form G) in importance, but that cannot be compared to the
national sample because health did not appear on Form D, which was
administered to them. The current sample ranked mature love ninth, whereas
the highest ranking in Rokeach’s three groups was fourteenth. In the opposite
direction, the current sample ranked national securityeighteenth, at the bottom

of the list, whereas the national sample ranked the value tenth, about in the

103

middle. A corollary value, a world at peace, was ranked first by the national
sample but only seventh by the current sample. Finally, the current sample
ranked pleasure somewhat more highly and a sense of accomplishment

somewhat less highly than did the national sample.

Table 4.5: Terminal value composite ranking comparison.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

value Rokeach 0-4 Rokeach 5—8 Rokeach Some l-l.S. Current Study
(n = 64) (n = 263) (n = 90) (N = 90)
A comfortable life 3 6 7 6
Equality 12 9 9 2
Exciting life 18 18 18 13
Family security 2 2 2 3
Freedom 4 3 3 5
Health NA NA NA 1
Inner harmony 9 13 13 15
Mature love 17 15 14 9
National security 10 10 10 18
Pleasure 14 16 17 11
Salvation 8 4 6 8
Self-respect 7 8 5 4
A sense of accomplishment 13 12 11 16
Social recognition 15 17 16 14
True friendship 6 7 12 10
Wisdom 11 11 8 12
A world at peace 1 1 1 7
Aworld of beauty 16 14 15 17

 

 

 

 

 

 

 

 

104

Table 4.6 contains the composite ranking comparison forthe instrumental
values. Both the national sample and the current sample ranked honesty, clean,

ambitious, and he/pfirl among the top five on the instrumental list.

Table 4.6: Instrumental value composite ranking comparison.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Value Rokeach 0-4 Rokeach 5-8 Rokeach Some HS. Current Study
(n = 64) (n = 263) (n = 320) (N = 90)
Ambitious 3 3 2 4
Broad-minded 10 7 8 12
Capable 11 13 11 8
Clean 2 4 7 1
Courageous 13 6 6 11
Forgiving 4 2 4 7
Helpful 7 5 5 5
Honest 1 1 1 2
Imaginative 18 18 18 17
Independent 15 14 14 10
Intellectual 16 16 16 16
Logical 17 17 17 15
Loving 6 10 10 3
Loyal NA NA NA 9
Obedient 14 15 15 18
Polite 12 11 13 13
Responsible 9 9 3 6
Self-controlled 8 12 12 14

 

 

 

 

 

 

 

105

Compared with the national sample, the current sample valued less highly
broad-minded, forgiving, and obedient. The current sample valued more highly
capable, independent, and loving. However, none of these differences was as
great as the differences found among the terminal values. Only four ranks
separated the current sample from the national sample in the cases of the widest
differences among the instrumental values (broad-minded and independent),
whereas eight ranks separated the two samples in the most-different terminal
values (eight ranks separated the groups regarding national security, and seven
ranks separated them in the case of equality).

Among the terminal values, there are two groups of differences of special
interest in the comparison of the current sample with the Rokeach national
sample. One is the lower rankings by the current sample of national securityand
a world at peace. These two values were ranked eight and six ranks lower,
respectively, by the current sample. This signals that, among this group, concern
regarding the possible imminence of war has declined over the past 20 years.

Asecond comparison of interest isthat linking the differences between the
current sample and Rokeach’s sample regarding the ranking of the values
exciting life, mature love, and pleasure. Each of these values was ranked more
highly by the current sample by about five ranks than by any of the three
Rokeach groups. This suggests that self-indulgence or immediate gratification
may have become more important to people without high school diplomas over

the years since Rokeach’s initial study.

106

Among the instrumental values there was one value with differences
between the current sample and Rokeach’s results that approached the
numerical distances evident in several terminal values. Lovingwas ranked third
by the current sample and sixth, tenth, and tenth by the three Rokeach groups.

To analyze more fullythe relationships between thethree Rokeach groups
and the current sample, Spearman rank correlation coefficients were calculated.
Health replaced another value on the terminal list after Rokeach’s administration
of the RVS, so an adjustment of the composite ranks was necessary. With this
value eliminated from the analysis, there were 17 values to study. Each value
affected by the elimination of health was then reassigned the appropriate new
rank order. These adjusted rank orders were then analyzed with the Spearman.
The researcher was interested in knowing how similar the current sample was
to each of the three non-high-school-completing groups in the Rokeach sample.

The results are shown in Table 4.7.

Table 4.7: Correlation of selected Rokeach groups with the current sample in
ranking terminal values.

 

II 0-4 Years Ed. 5-8 Years Ed. Some High School II
II Spearman rho .55 .63 .65 ll

 

 

 

 

 

As can be seen in Table 4.7, the correlation in rankings between the group

with the fewest years of education in Rokeach’s sample and the current sample

107

was the lowest, at rho = .55. The two groups with more years of education had
higher correlations (rho = .63 and .65, respectively) with the current sample.
A similar analysis of the instrumental values was conducted. From this list
of 18 values, loyalhad replaced another value since the Rokeach administration.
Composite rankings were then adjusted for this list, resulting in 17 values to be
analyzed. The question of interest again was how the Rokeach groups

compared with the current sample. The results are shown in Table 4.8.

Table 4.8: Correlation of selected Rokeach groups with the current sample in
ranking instrumental values.

 

II 04 Years Ed. 58 Years Ed. Some H.S. II
II Spearman rho .86 .80 .81 II

 

 

 

 

 

The correlations between each of the three Rokeach groups and the
current sample were at rho = .80 or greater for the instrumental values. The
highest correlation was that between the group with the fewest years of
education in Rokeach’s sample and the current sample.

This analysis confirmed the earlier observation that differences between
the 1974 sample and the current sample were more pronounced among the
terminal values than the instrumental values. Applying Rokeach’s means—ends
conceptualization, there is support for the argument that, among high school

dropouts, in 20 years there has been little change in the perception of how to

108
achieve or advance, and more change in the perception of which goals should
be pursued.

These data also indicate that there was substantial agreement in values
between the Rokeach groups and the current sample. The correlations support
the argument that, among nonholders of high school diplomas, there has not
been a wholesale change in the perception of values in the past 20 years. The
differences that do exist, a lessening of concern with global peace and an
increased interest in or valuing of hedonism, are encapsulated in a general trend
of similarity in the rankings of other values.

In sum, the national sample and the current sample had some differences
in the way they ranked the values. Seven terminal and six instrumental values
were separated by three or more ranks when comparing the current absence-
rate groups. The greatest differences were found among the terminal values.
Overall, the current sample could not be readily distinguished from the national

sample in the cases of 23 of the 36 values on the RVS.

onmhesisn

Hypothesisl There is no relationship between parental value rankings
and their children’s school attendance patterns.

All parental rankings (N = 90) of each of the 36 values were analyzed
against the ranking of each student’s days of absence, matching the particular

parent with the particular student. The Spearman rank correlation coefficient, or

109
rho, was the appropriate test for this question (Siegel, 1956, p. 202). Table 4.9

shows the results for the terminal values.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 4.9: Correlation of terminal value rankings with attendance rates.
Value Spearman Rho Significance
A comfortable life -.5579 .000"
Equality -.11 17 .295
Exciting life -.4221 .000"
Family security -.1636 .123
Freedom -.0912 .393
Health .0140 .896
Inner harmony .0220 .837
Mature love -.0618 .563
National security .1 171 .272
Pleasure -.4122 .000"
Salvation -.2581 .014
Self-respect .2463 .019
A sense of accomplishment .5500 .000"
Social recognition . 1874 .077
True friendship -.0682 .523
Wisdom .5436 .000"
A world at peace .2981 .004"
A world of beauty .1120 .293

 

"Significant at the .01 level.

 

110

In interpreting the table, the comparison is between varying levels of
parental rankings and varying numbers of days of student attendance (as
represented by days of absence). A positive correlation means that as parents
ranked the value more highly, their children were more often in school. A
negative correlation means that as parents ranked the value more highly, the
children were less often in school.

There were 12 terminal values for which no correlations were found
between the way parents ranked the value and the school attendance record of
the children. The low correlations among this group of 12 values, ranging from
absolute values of .06 (mature love) to .25 (salvation), were not significant at the
.01 level. In fact. of these 12 values, 9 had correlations of .16 or less. This
suggests that, for half of the terminal values, there was no relationship between
the way parents ranked the values and the level of student school attendance.

However, variations in the parental rankings of six terminal values were
found to correlate with student attendance rates. The highest correlations were
for a comfortable life (-.56), a sense ofaccomp/ishment (.55), and wisdom (.54).
The next highest correlations between parental rankings and student school
attendance were for an exciting life (-.42) and pleasure (-.41). The lowest
correlation of significance was for a worid atpeace (.30). There were no other
value rankings that related to school attendance rates at the .01 level of

significance or less.

111

Of these six values, all appeared to correlate in a linear fashion. There
were, however, two different linear directions. For three values (a sense of
accomplishment, wisdom, and a world at peace), student attendance increased
as the parental ranking of the value increased. For three other values (a
comfortable life, an exciting life, and pleasure), student attendance dropped as
parental ranking increased.

This finding was a preliminary indication of whether a particular pattern of
parental values was related to student school attendance. The direction and
magnitude of the correlations for these values are shown in Table 4.10. These
results were particularly intriguing, given the theoretical basis for the study and

the research questions.

Table 4.10: Terminal values correlated with student attendance.

 

 

 

 

 

Higher Parent Ranking, Higher Parent Ranking,

Increased School Attendance Decreased School Attendance
A sense of aCcomp/ishment A comfortable life

(.55) (~56)

Wisdom An exciting life
(.54) (-.42)
A world at peace Pleasure
(.30) (-.41)

 

 

 

The three terminal values correlating in a positive direction with students
attendance, as shown in Table 4.10, were future-oriented. The three terminal

values correlating in a negative direction (as parents ranked these higher, their

112

children were less often in school) were present-oriented. The magnitudes ofthe
relationships were similar in both directions and were sufficiently high to suggest
moderately strong relationships.

Among the instrumental values, 1 1 values did not relate to student school
attendance. As with the noncorrelated terminal values, the correlations for these
11 instrumental values were low and not significant at the .01 level. Seven of
these 11 values had correlations of .16 or less, suggesting weak if not random
relationships to school attendance. Examples of these values are honest,
helpful, capable, and polite. Table 4.11 shows the results of the Spearman
analysis for the instrumental values.

But for seven instrumental values, correlations were discovered with
student school attendance. The two strongest correlations were for logical (.47)
and imaginative (.45). The next most strongly related values were intellectual
(.38), courageous (-.35), and ambitious (.35). The final value that related to
school attendance at a level of significance less than .01 was forgiving (.28).

Amongtheinstrumental values correlated with student school attendance,
five (logical, imaginative, intellectual ambitious, and forgiving) displayed positive
correlation coefficients. As parents ranked these values more highly, their
children were more often in school. Two values (courageous and self-controlled)
had negative coefficients, meaning that parents who ranked these values more

highly had children who were less often in school.

113

Table 4.11: Correlation of instrumental value rankings with attendance rates.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Value Spearman Rho Significance
Ambitious .3458 .001 "
Broad-minded -.0685 .521
Capable .0375 .725
Clean -.1249 .241
Courageous -.3621 .000"
Forgiving .2843 .007"
Helpful -.1944 .066
Honest .0663 .535
Imaginative .4501 .000"
Independent -.0892 .403
Intellectual .3766 .000"
Logical .4708 .000"
Loving .0880 .410
Loyal -.2023 .056
Obedient -.1703 .109
Polite -.0541 .612
Responsible .2102 .047
Self—controlled -.3540 .001 "

 

 

 

 

 

"Significant at the .01 level.

As with the results reported for the terminal values, these results for the
instrumental values provided additional insight into the questions of interest.
Table 4.12 contains details about the direction and magnitude for each

instrumental value found to relate to student school attendance.

114

Table 4.12: Instrumental values correlated with student attendance.

 

 

 

 

 

 

 

Higher Parent Ranking, Higher Parent Ranking,
Increased School Attendance Decreased School Attendance
Logical Courageous

(.47) (-.36)

Imaginative Self-controlled
(.45) (-.35)

Intellectual
(.38)

Ambitious
(.35)

Forgiving
(.28)

 

 

 

Being logical and being imaginative were increasingly important to parents
whose children attended school with increasing regularity. So were being
intellectual, ambitious, and forgiving. For parents whose children missed more
and more school, being courageous and being self-controlled were increasingly
important. In comparison with the terminal values, the magnitudes of the
relationships found among the instrumental values were not quite as large.
There were no .5 correlations among the instrumental values, whereas there
were three correlations of that magnitude among the terminal values. This
finding suggests that the way parents rank the terminal values might be a better
overall indicator of their children’s school attendance patterns than the parental

ranking of the instrumental values.

115

Considering all 36 values on the Rokeach Value Survey, the way parents
ranked 23 values showed no relationship to their children’s school attendance
rates. However, the parental rankings of 13 values did relate to differences in
their children’s school attendance rates. Eight of the correlations were in the
direction that, as parents ranked the values more highly, their children were more
frequently in school. Five of the correlations were in the opposite direction, such
that, as parents ranked those values more highly, their children were less often
in school. Hypothesis 1 was confirmed for these 13 values, 6 of which were

terminal values and 7 of which were instrumental values.

Hypothesisz

Hypothesifl: When grouped on the basis of their children’s school
attendance patterns, there are no differences in parental value rankings.

The results reported forthe first hypothesis suggest that there were some
relationships between the rankings parents assigned to certain values and their
children’s school attendance patterns. The second hypothesis was formulated
to investigate whether differences in value rankings existed when the parents
were grouped according to their children’s school attendance rates.

The 90 parent respondents were sorted into three groups on the basis of
their children’s days of absence from school. The Kruskal-Wallis one-way
ANOVA was used to determine whether the groups differed in ranking the
values. Results of the analysis are presented in Tables 4.13 and 4.14. The

terminal values are shown in Table 4.13, and the instrumental values appear in

116

Table 4.14. The .01 level was established as the criterion for statistical

significance.

Table 4.13: Results of the Kruskal-Wallis ANOVA: Terminal values.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Value Mean Rank Mean Rank Mean Rank Chi- Signif.
Low-Absence Med-Absence High Absence Square

A comfortable life 11.28 8.33 4.68 32.27 .0000"
Equality 7.66 6.30 6.21 3.90 .1422
Exciting life 13.34 9.97 7.43 23.91 .0000"
Family security 8.13 6.53 6.75 1.08 .5825
Freedom 8.88 7.03 7.96 3.26 .1962
Health 6.78 5.50 7.21 3.69 .1583
Inner harmony 10.59 11.83 10.68 1.28 .5283
Mature love 10.81 8.23 9.82 4.02 .1342
National security 13. 59 15.27 14.32 2.98 .2255
Pleasure 13.34 8.97 7.61 22.45 .0000"
Salvation 11.00 7.87 8.39 5.83 .0541
Self-respect 5.66 8.93 7.71 7.18 .0276
A sense of accomplish- 8.28 11.00 14.39 31.00 .0000"
ment
Social recognition 9.91 9.87 12.93 4.82 .0890
True friendship 8.63 12.33 8.39 9.48 .0087"
Wisdom 7.16 10.87 13.50 35.13 .0000"
A worid at peace 5.72 10.93 10.25 14.28 .0008"
A world of beauty 10.31 11.20 12.71 1.03 .5988

 

"Significant at the .01 level.

117

Table 4.14: Results of the KruskaI-Wallis ANOVA: Instrumental values.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Value Mean Rank Mean Rank Mean Rank Chi- Signif.
Low-Absence Med-Absence High-Absence Square

Ambitious 5.69 8.27 8.54 9.69 .0079"
Broad-minded 10.59 8.57 10.36 3.37 .1859
Capable 8.84 8.97 8.75 0.20 .9059
Clean 7.16 6.77 5.00 1.62 .4440
Courageous 11.84 9.27 7.82 11.92 .0026"
Forgiving 7.03 8.40 10.61 9.23 .0099"
Helpful 10.78 7.33 6.89 9.63 .0081"
Honest 5.75 7.57 5.79 3.60 .1653
Imaginative 11.75 12.60 15.61 19.93 .0000"
Independent 10.25 9.10 8.82 1.53 .4665
Intellectual 10.63 13.20 14.29 18.00 .0001 "
Logical 10.19 12.90 14.00 21.62 .0000"
Loving 5.91 6.57 7.32 0.93 .6295
Loyal 10.75 9.20 8.07 4.68 .0964
Obedient 14.19 13.70 11.82 4.42 .1098
Polite 10.06 10.23 9.25 0.72 .6967
Responsible 6.50 9.07 9.96 6.25 .0438
Self-controlled 13.13 9.30 8.11 11.13 .0038"

 

 

 

 

 

 

"Significant at the .01 level.

 

118

As shown in Table 4.13, there were seven terminal values that were
ranked differently by the three parent groups. A comfortable life, an exciting life,
pleasure, a sense of accomplishment, true friendship, wisdom, and a world at
peaceall showed statistically significant differences. Among these seven values,
three different patterns were observed as the parent groups ranked the values.
In some cases, the parent group rankings appeared linear, having stepwise
increases in rankings. In other cases, the rankings appeared linear, but
decreasing among the groups. Finally, other relationships appeared to be
curvilinear, evidencing no linear pattern.

These relationships for these seven terminal values are presented in
Figure 4.1. Note that an increased ranking for a parent group is represented by
a larger bar on the chart, in that a ranking of 1 is high, or most important, on the
Rokeach Value Survey. Similarly, a decreasing ranking is represented by a
smaller bar because a ranking of 18 is low, or least important.

The values ranked less important bythe parents of low-absence students,
more important bythe parents of medium-absence students, and most important
by the parents of high-absence students are shown on the left column of Figure
4.1. These values, a comfortable life, pleasure, and an exciting life,
demonstrated an apparent linear increase in ranking as student school
attendance dropped. Parents who ranked these values more highly had children

who were less often in school.

119

 

 

 

 

    

 

 

 

 

    

 

 

 

    
 

; 1 A comfortable life i A sense of accomplishment
: J ‘
4
s " 5
‘ 0
s. E : x I
C -
S 1.0 - 5 ‘.°
. II
5 5%: E I:
E E :z
e 1:: g ..
5 e i;
o I. 4 0
Low Medium High Low ium High
Absence group Absence group
I]
a : Pleasure
; -
' I 7 61
x Z . 8.97 , , sé
E .21 e
= II : C
8 ii. 8
E 1;: E
D. _ D.
a 52‘ 5
cf ‘ 0
Low Medium High Low Medium High
Absence group Absence group
I 1
Z 2
3 An exciting life 2 True friendship
5
0 0
7 7
'EE : 9.97 'E I
S E in
C ‘ C II
«a cc 1:
e e 1:
g. g- I:
g . 5 :z ,. .
Low Medium High Low Medium High
Absence group Absence group
I
2
3 A world at peace
5 .72
6
7
+5 I
S 10
g I;
a) I:
g I:
3 I:
a ll
Low Medium High
Absence group

 

 

Figure 4.1: Parental mean rankings, by absence groups, for terminal values

ranked differently as tested by the Kruskal-Wallis.

120

The two values ranked in the opposite direction are shown at the top of
the second column. A sense ofaccomp/ishment and wisdom both displayed an
apparent linear progression of being ranked less important by parents as their
students’ school attendance dropped. Stated conversely, these two values
seemed to increase in importance to parents as their children attended school
more frequently.

Two other values appeared to display the third pattern, that of being
curvilinear. True friendship, shown by the Kruskal-Wallis to be ranked differently
among the parent groups, was ranked as less important by the parents of
medium-absence students, and as more important by the parents of both low-
and high-absence students. A world atpeace was seen as more important by
the parents of low-absence students, but as less important by the parents of the
other two groups.

The magnitude of the differences in rankings by the three parent groups
are also of interest. On Figure 4.1 it can be seen that the low-absence parent
group’s mean ranking of a comfortable life was lower than the mean rankings of
the other two parent groups. A mean ranking OH 1.28 for the low-absence group
is lower than the mean ranking of 8.33 for the medium-absence group, which in

turn is lower than the mean ranking of the high-absence group (4.68).

121

These relationships are contrasted with those for true friendship, for
example. Note that the bars forthe low-absence group (mean rank 8.63) and the
high-absence group (mean rank 8.39) are similar in size.

Thus, Figure 4.1 also illustrates that, for some terminal values found to be
ranked differently by the parent groups, the magnitude of the differences among
the groups was greater (note a comfortable life, pleasure, an exciting life, a
sense of accomplishment, and, in particular, wisdom). For other values, the
magnitude of the differences was not so great (see true friendship and a world
atpeace).

A similar analysis was conducted for the instrumental values that the
Kruskal-Wallis had indicated parent groups had ranked differently (see Table
4.14). Eight values showed statistically significant differences in the way the
parent groups ranked them. Those instrumental values were ambitious,
courageous, forgiving, he/pflrl, imaginative, intellectual, logical, and self-
controlled

As with the terminal values, the patterns of positive and negative linearity
were evident in examining the differences among the parent groups in the way
they ranked the instrumental values. Forthe values he/pfirl, self-controlled, and
courageous, parents ranked each as being more important as their children
missed more and more school. For ambitious, intellectual, logical, imaginative,
and forgiving, parents ranked each as being less important as their children

missed more and more school. However, the magnitude of the differences

122
became problematic in interpreting this analysis for the instrumental values.
Figure 4.2 is helpful in describing these results.

From Figure 4.2, it appears that the trend for the value helpfu/was linear
and positive, such that as student absence increased, parental ranking of the
value increased. Yet the magnitude of the difference between the mean rank of
the medium-absence parent group (7.33) and that of the high-absence group
(6.89) was not great. Thus, the question was raised as to whether the
relationship was linear or curvilinear.

This problem is an artifact of the use of the Kruskal-Wallis test. The test
was valuable in that it could be used to analyze the three parent groups jointly.
Statistically, this pooled the variance. due to error for all three groups, thus
increasing the confidence that any differences found were not attributable to
chance. However, it did not allow for conclusions to be drawn regarding
differences between any pair of groups.

When comparing any pair of groups in this analysis, error variance due to
the absence of the third group could not be calculated. A typical adjustment is
to change the alpha level on a follow-up test in order to be more certain that
results are not due to chance. This has the effect of partitioning the variance,
even though that is not precisely what has been done. In the case of comparing
these three groups, dividing the alpha level by three would provide the required

standard. However, when the Mann-Whitney was used to compare pairs of

123

 

      

; Helpful l Intellectual
I I
E ; 7.33 "é :
E ' g I:
E e I:
a s
<5 5 ll -
Low Medium High Low Medium High
Absence group Absence group

 

Self-controlled Logical

5:3I-iuuhufl—

      

Group mean rank
Group mean rank

Low Medium High

 

   

 

   

 

Medium
Absence group Absence group
i Courageous é Imaginative
.. : s
' x
E i 9.27 E Z
c ' c .1,
I“ I m in
a) a) II
E E
3 s
0 a
Low Medium High Low Medium High
Absence group Absence group
i Ambitious x i Forgiving
x a C s
g E 8.27 E 2 7'03 8.40
g. i t i:
12 a- :2
3 is h is
2 u (D u
<5 II II , ,
Low Medium High Low Medium High
Absence group Absence group

 

 

 

 

Figure 4.2: Parental mean rankings, by absence groups, for instrumental
values ranked differently as tested by the KruskaI-Wallis.

124

groups, the significance values were so low (ranging from .0000 to .0099) that
there would have been little practical significance in using the procedure.

When the Mann-Whitney test was used to compare the groups in this
pairwise fashion, the intention was to determine to what the aggregate
differences found through administration of the KruskaI-Wallis could be
attributed. It was possible that two parent groups were quite similar to each other
and quite different from the third group as they ranked the values (the curvilinear
model). Or it could have been that all groups were different from each other (the
linear model). The Mann-Whitney was performed on each of the 15 values, 7
terminal and 8 instrumental, which had been found to be ranked differently under
the KruskaI-Wallis. The results of the Mann-Whitney are shown in Tables 4.15

and 4.16.

Table 4.15: Results of the Mann-Whitney test: Terminal values.

 

Value

Low-Absence vs.
Medium Absence

Medium-Absence
vs. High-Absence

Low-Absence vs.
High-Absence

 

A comfortable life

+

+

+

 

Exciting life

+

+

 

Pleasure

+

+

 

A sense of
accomplishment

 

True friendship

 

Wisdom

 

 

A world at peace

 

 

 

 

Note; + denotes values on which significant differences were found.

 

125

Table 4.16: Results of the Mann-Whitney test: Instrumental values.

 

Value

Low-Absence vs.
Medium-Absence

Medium-Absence
vs. High-Absence

Low-Absence vs.
High-Absence

 

Ambitious

+

.1.

 

Courageous

4.

+

 

Forgiving

+

 

Helpful

 

Imaginative

 

Intellectual

 

Logical

 

Self-controlled

 

 

 

 

 

Note: + denotes values on which significant differences were found.

From Table 4.15 it can be seen that a comfortable life was ranked

differently in all three comparisons. The mean ranking increased about four

ranks in comparing the low-absence (11.28) and medium-absence (8.33) groups,

and it increased another four ranks in comparing the medium-absence (8.33) and

high-absence (4.68) parent groups. A lower value means a higher ranking, with

a rank of 1 being the highest or most important value and a rank of 18 being the

lowest or least important value. Thus, a comfortable lifewas ranked significantly

differently by each parent group, and the mean ranking increased as the

children’s school attendance decreased. As parents thought a comfortable life

was more important, their children were less likely to be in school.

 

126

An exciting life, pleasure, and a world atpeace were ranked differently by
the low-absence group when compared with both the medium-absence and high-
absence groups, but no difference was found in comparing the mean rankings
of the medium- and high-absence groups. For an exciting life, the low-absence
group’s mean ranking of 13.34 was significantly lower than the mean ranking of
either the medium-absence (9.97) or high-absence (7.43) group. Again, a
difference of about four ranks in the mean ranking statistic was a significant
difference. Yet the 2.5 ranks separating the medium- and high-absence groups
was not a significant difference. Pleasure displayed a nearly identical pattern to
an exciting life in the way it was ranked by the parent groups, as the mean ranks
were very similar (the low-absence group mean rankings were tied for pleasure
and an exciting life at a mean ranking of 13.34 for each). A world at peace,
althoughonce again seen differently by the low-absence group but similarly by
the medium- and high-absence groups, was ranked much higher (being more
important) by the low-absence group. More than five mean ranks separated the
Iow- from the medium- and high-absence groups for this value.

A sense of accomplishment was ranked differently in all comparisons.
The relationship was linear; as student attendance decreased, so did parental
rankings. The low-absence mean ranking of 8.28 was higher than the medium-
absence mean ranking of 1 1.00, which in turn was higher than the high-absence
mean ranking of 14.34. Parents who thought this value was important had

children who were more often in school. The terminal value wisdom had the

127
same pattern, with the low-absence mean ranking (7.16) higher than the
medium-absence mean ranking (10.87), which in turn was higher than the high-
absence mean ranking (13.50).

True fifendshipexhibited a ranking pattern unlike that of the otherterrninal
values. For this value, the differences between the mean ranking by the low-
absence (8.63) and high-absence (8.39) groups were insignificant. But both of
these groups ranked the value as more important than did the medium-absence
group (12.33).

Turning to the instrumental values (Table 4.16), it is first noted that there
were no instrumental values for which significant differences were found among
all three parent groups. For all eight tested instrumental values, a significant
difference was found between the mean ranking of the low-absence group when
compared with the mean ranking of the high-absence group. For six of the
values (ambitious, courageous, helpful, intellectual, logical, and self-control/eo),
differences were also found between the mean rankings of the low- and medium-
absence groups. For one value (imaginative), a difference in mean rankings was
found between the medium- and high-absence groups. Thus, for the
instrumental values, none of the relationships that appeared to be linear when
observing Figure 4.2 were found to be so. Most of the differences among the
parental mean rankings for the instrumental values were found under the Mann-
Whitney to be differences between the low-absence parent group and the other

two groups, the medium- and high-absence groups.

128

For he/pfirl, self-controlled, and courageous, the low-absence group mean
rankings revealed that these parents thought these values were less important
than did the other two parent groups. in general, the differences for these three
values were about three ranks separating the low- from the medium- and high-
absence groups.

For ambitious, intellectual, logical, and forgiving, the low-absence parent
group mean ranking was higher than that of the other groups, meaning that the
low-absence parents thought these values were more important than did the
other groups. Logica/and intellectual displayed nearly identical profiles. For all
four of these values, the magnitude of the differences in mean rankings was
again generally on the order of three ranks separating the groups.

Imaginative was the anomaly among the instrumental values. This was
the one value the low- and medium-absence parents ranked similarly (mean
ranking of11.75 by the low-absence parents and 12.60 by the medium-absence
parents). The high-absence parents (mean ranking 15.61) saw this value as
much less important than did the other two groups.

Considering both terminal and instrumental values, the use of the Mann-
Whitney revealed that the majority of differences in the way the three parent
groups ranked the values resided in the differences between the low-absence
group when compared with the other two groups. Three values, a comfortable
life, asense ofaccomp/ishment, and wisdom, all terminal values, were confirmed

as having linear relationships across the parent groups. For a comfortable life,

129

as parents thought the value was more important, their children were less often
in school. As parents assessed a sense ofacoomplishmentand wisdom as more
important, their children were more often in school. Hypothesis 2 was rejected

for the 15 values, 7 terminal and 8 instrumental, as described above.

Hypothesis}

Hypothesis}: The rankings of low-absence parents (Rm) for a sense of

accomplishment, wisdom, imaginative, intellectual, logical, mature love,

responsible, a comfortable life, pleasure, salvation, clean, and obedient

(educational values) will be the same as those of the medium-absence

(RM) and high absence (RW) parents.

There were three tests for this hypothesis. The Spearman, the Kruskal-
Wallis, and the Mann-Whitney each provided some indication as to whether the
values in the 12-value model of educational predisposition were associated with
group differences and school absence behavior. The results from each of these
tests on the terminal values are shown in Table 4.17.

Six terminal values were predicted to be related to school-attendance
behavior by the logic 0f the 12-value model. The model was based on
Rokeach’s (1974) findings relative to values that discriminated among
respondents with varying years of education (see Chapter I). The predicted
terminal values, as shown in Table 4.17, were a comfortable life, mature love,
pleasure, salvation, a sense of accomplishment, and wisdom.

As shown in Table 4.17, variations in the parental rankings for a

comfortable life, a sense of accomplishment, and wisdom were correlated with

130

Table 4.17: Summary of results—terminal values.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Value Predicted by Spearman Kruskal- Mann- Mann- Mann-

12-Value Rank Wallis Whitney Whitney Whitney
Model? Correl. Low/Med. Med/High Low/High

A comfortable Yes -. 56 + + + 4-

life

Equality

Exciting life -,42 + + +

Family security

Freedom

Health

lnnerhannony

Mature love Yes

National security

Pleasure Yes -.41 + + +

Salvation Yes -.26

Self-respect

A sense of Yes .55 + + + +

accomplishment

Social

recognition

True friendship + + +

Wisdom Yes .54 + + + +

A world at peace .30 + + +

A world of

beauty

 

 

 

 

 

 

 

bible: + denotes value on which significant differences were found.

 

131

children’s school attendance, with coefficients of -.56, .55, and .54, respectively.
Each of these values was also ranked differently by the parents when grouped
according to their children’s school attendance rates as tested by the Kruskal-
Wallis. Further, one parent group ranked each value differently from the other
two parent groups as tested by the Mann-Whitney.

In the case of both a sense of accomplishment and wisdom, parents
whose children were in school more often ranked these values as more
important, which was in the predicted direction. The mean ranking for a sense
ofaccomp/ishmentforthe low-absence group was 8.28, forthe medium-absence
group 11.00, and for the high-absence group 14.39. This means that the high-
absence group ranked six other values as more important than a sense of
accomplishment when compared with the parents of low-absence students.
Similar rankings existed for wisdom, as reported above.

In the case of a comfortable life, the relationship was opposite; parents
whose children were in school more often ranked this value as less important,
which again was in the predicted direction. The magnitude of differences here
was somewhat larger, with the low-, middle-, and high-absence groups’ mean
rankings calculated at 11.28, 8.33, and 4.68, respectively.

For the values an exciting life, pleasure, and a wortd at peace, the rank
correlation coefficients were -.42, -.41, and .30, respectively, so each had a
relationship with student school attendance. These three values were found by

the use of the Kruskal-Wallis to be ranked differently by the parent groups. Use

132

of the follow-up Mann-Whitney revealed that differences existed in the mean
rankings of these values when comparing the low-absence group with both the
medium- and high-absence groups, but not when comparing the medium- and
high-absence groups. The direction of differences was the same for an exciting
life and pleasure (as predicted), in that parents whose children were in school
more often ranked these values as less important. Parents whose children were
in school more often ranked a won'datpeace as being more important. It should
be noted that neither an exciting life nor a won'd at peace was predicted by the
logic of the 12-value model to be related to school-attendance behavior.

Salvation was related to school attendance, with a rank correlation
coefficient of -.26, in the predicted direction. Parents whose children were more
often in school ranked this value as being less important. However, this
moderate correlation was not accompanied by any detected differences in the
ways the parent groups ranked the value.

Variance in the parental rankings of true friendship was not related to
school-attendance patterns of their children. However, as shown in Table 4.17,
there were differences in the ways parents ranked the value as found under the
Kruskal-Wallis. This can be attributed to differences in the rankings when
comparing the Iow- and medium-absence parents, and when comparing the
medium- and high-absence parents. The low- and high-absence parent groups’
mean rankings were 8.63 and 8.39, respectively. The medium-absence parents

ranked this value lower than did the other two groups, with a mean ranking of

133

12.33. True fitendship was not one of the values predicted by the 12-value
model.

Variations in parental rankings for mature love, although predicted by the
logic of the 12-value model to correlate with school attendance, did not
demonstrate such a relationship. Nor did any of the parent groups rank it
differently from one another.

The summary of results for the instrumental values is presented in Table
4.18.

Six instrumental values were predicted to be related to school attendance:
clean, imaginative, intellectual, logical, obedient, and responsible. For three
values, clean, obedient, and responsible, no relationships were discovered under
any of the tests.

Forthree other values, imaginative, intellectual, and logical, relationships
were found in the predicted directions. The correlation coefficients for these
values were .45 (imaginative), .38 (intellectual), and .47 (logical). Parents whose
children were in school more often ranked these values as more important than
did parents whose children missed school more often. In the case of
imaginative, the differences in mean rankings between the medium- and high-
absence groups, and between the Iow- and high-absence groups, were
significant. In the cases of both intellectua/and logical, the differences between

the low- and medium—absence groups, and between the low- and high-absence

Table 4.18: Summary of results—instrumental values.

134

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Value Predicted Spearman Kruskal- Mann- Mann- Mann-
by 12-Value Rank Wallis Whitney Whitney Whitney
Model? Correl. Low/Med. Med/High Low/High
Ambitious .35 + + +
Broad-minded
Capable
Clean Yes
Courageous -. 36 + + +
Forgiving .28 + +
Helpful + + +
Honest
Imaginative Yes .45 + + +
Independent
Intellectual Yes .38 + + +
Logical Yes .47 + + +
Loving
Loyal
Obedient Yes
Polite
Responsible Yes
Self-controlled -.35 + + +

 

 

 

 

 

 

 

Note: + denotes value on which significant differences were found.

 

135

groups, were significant. In all three cases, the mean rankings for the low-,
medium-, and high-absence groups were about 10, 13, and 14, respectively.

Among the nonpredicted values, self-controlled, ambitious, and
courageouswere found to exhibit similar patterns. All had correlation coefficients
of about .35. The correlations for self-controlled and courageous were negative,
meaning that parents who ranked these values as more important had children
who missed school often. For ambitious, the correlation was positive, meaning
that parents who ranked this value as more important had children who were in
school more often. In the comparisons between the low-absence group and both
of the other groups, the parental mean rankings were significantly different (but
not when comparing the medium- and high-absence parent groups) for these
three values.

The value forgiving, another nonpredicted value, had a correlation
coefficient of .28, indicating that parents who ranked this value more highly had
children who were more likely to be in school more often. Although found to be
ranked differently by the parent groups using the Kruskal-Wallis, the follow-up of
the Mann-Whitney revealed that only the Iow- and high-absence groups were
significantly different (mean ranks of 7.03 and 10.61, respectively) in the ways
they ranked this value.

Finally, ranking of the value he/pfirl, although not related to children’s
school attendance, was different for some of the parent groups. The low- and

medium-absence groups ranked it differently (mean ranks of 10.78 and 7.33,

136
respectively). So did the low- and high-absence groups (mean ranks of 10.78
and 6.89, respectively).

Considering the 36 values on the Rokeach Value Survey, higher parental
rankings for two terminal values were predicted and found to have comparatively
strong relationships to increased student attendance: a sense of
accomplishment and wisdom. Three values were also predicted and found to
have this pattern. Those values were imaginative, intellectual, and logical.

Two other terminal values were predicted and found as having the
opposite pattern, that being a higher ranking by the parent and decreased school
attendance by the child, and a comparatively strong relationship. Those values
were a comfortable life and pleasure. Another terminal value, salvation, was
predicted to have this pattern, but it was found to have only a slight relationship.
No instrumental values were both predicted and found to have this direction of
a relationship.

Several nonpredicted values did evidence a relationship to school
attendance. The ranking of an exciting life (terminal list) and courageous and
self-controlled (instrumental list) were related to school attendance in the
negative pattern (higher ranking by parent, less school attendance by the
student).

The opposite relationship existed for the nonpredicted values a world at

peace (terminal list) and ambitious and forgiving (instrumental list). As parents

137

ranked these values as being more important, their children were more often in
schooL

The 12-value model was confirmed in the cases of five of six predicted
terminal values (a comfortable life, pleasure, salvation, a sense of
accomplishment, and wisdom), and in the cases of three of six predicted
instrumental values (imaginative, intellectual, and logical). Research results also
illuminated four other terminal values and five instrumental values not included

in the model that exhibited relationships similar to those predicted by the model.

Summant

Initial findings and results related to the three hypotheses were described
in the four preceding sections. Nearly all values had been ranked either first or
last by at least 1 of the 90 respondents, and standard deviations for mean
rankings ranged about three points for both lists of values. Differences between
the current sample and the Rokeach national sample (nonholders of high school
diplomas) were found in the rankings of 13 of the 36 values, with the greatest
differences among the terminal values.

The first null hypothesis was rejected for 15 values regarding the
association between parental ranking and student days of school absence.
Three correlation coefficients exceeded the .5 level, four were between .4 and
.5, and five were between .3 and .4. A single value ranking correlated with
school absence with a coefficient of .28. The majority of the highest coefficients

were among the terminal values.

 

138

In comparing groups of parents sorted on the basis of their children’s
absence rates, the null hypothesis was rejected for 15 values, 7 among the
terminal and 8 among the instrumental values. In a follow-up analysis, three
terminal values (but no instrumental values) were demonstrated to have been
ranked differently in all comparisons ofthe three parent groups, fourterminal and
six instrumental values were ranked differently in two of the three parent-group
comparisons, and one instrumental value was ranked differently in one of the
three parent-group comparisons.

In examining the 12-value model of educational predisposition, the null
hypothesis was rejected for five of six terminal values and for three of six
instrumental values. The evidence was a combination of the results presented

for the first two hypotheses.

CHAPTER V

SUMMARY, IMPLICATIONS OF THE FINDINGS, RECOMMENDATIONS

FOR FURTHER RESEARCH, AND CONCLUSION

This chapter begins with a summary of the study and a review and
discussion of the major findings. The implications of the findings and
recommendations for further research are presented next. The chapter ends

with a section concluding the study.

Summau

The theories of Tonnies, Durkheim, Parsons, and Rokeach suggest that
social/institutional or commonly held values are interdependent with personal or
individually held values. Parsons and Rokeach, in particular, discussed the idea
that values are antecedents to behavior. A review of literature related to school
absence revealed that the problem of school absenteeism has been pervasive
across time and cultures. No studies were found that addressed the possible
relationship between parents’ value patterns and their children’s school absence
patterns. Some investigators have found that early school absenteeism had a
serious negative influence on subsequent school achievement, adult income,

and other life circumstances.

139

140

The perspective of the family as it contributes to students’ academic life,
a line of investigation developed in research into high school dropping-out
behavior (Okey, 1990) and high school achievement (Clark, 1983) was adopted
for the current study. The researcher’s purpose in this study was to identify
which parental values (representing the family perspective), if any, related to the
school-absence behavior of children.

Five research questions guided this study. Three hypotheses were
formulated to investigate those questions. The study was conducted in a
midwestern urban school district enrolling 20,000 students in kindergarten
through grade 12. A random sample of children enrolled in kindergarten through
grade 3 was selected from among families in which the mother had not
completed high school. The sample consisted of 226 children from the
population of 1,604 cases. Rokeach Value Surveys were mailed to the parents
of these children. The response rate was 39.8%, reflecting 90 survey returns.

Parents’ rankings of the values on the survey were analyzed using SPSS
on the mainframe computer at Michigan State University. The rankings were
compared to those of Rokeach’s (1973) non-high-schooI-graduate groups.
Subsequently, the Spearman rho, or rank correlation coefficient, was used to
determine whetherthe current parent rankings of the values were related to their
children’s school-absence rate. The KruskaI-Wallis one-way ANOVA was then
employed to examine parent rankings for differences between groups when

parents were sorted according to their children’s absence rate (low, medium, or

141

high). The Mann-Whitney Utest was used as a follow-up to this analysis. A 12-
value model of educational predisposition was developed, based on results
obtained by Rokeach (1973). The three nonparametric statistical analyses were
then used to examine the accuracy of the model developed for this study.

The analyses revealed that some differences did exist between the non-
high-school graduates surveyed by Rokeach and those in the current study.
Further, some parental value rankings were correlated with students’ absence
rates. Also, when grouped on the basis of their children’s absence rates, the
parent groups differed significantly in their patterns of value rankings. Finally, the

12-value model was partially confirmed.

vi w ' i ' r ' i

The researcher’s purpose in this study was to investigate the possible
relationship between parental values and elementary students’ school
absenteeism. It was thought that a description and explanation of such a
relationship would illuminate one facet of a problem that has persisted overtime
and across cultures.

The first two research questions focused on describing the way members
of the current sample ranked the values and comparing these rankings to those
of other samples. The first two questions were:

1. How did parents in the current sample rank the values on the

survey?

142

2. How do these results compare to those from other samples?

The terminal values ranked most highly by the current sample were
health, equality, and family security. The instrumental values ranked at the top
were clean, honest, and loving. The terminal values receiving the lowest
rankings by the current sample were a sense of accomplishment, a world of
beauty, and national security. The instrumental values ranked lowest were
intellectual, imaginative, and obedient.

Nearly every value on both lists was ranked first or last by at least one
respondent, and the standard deviations of the mean rankings for each value
were often similar. In the cases of the terminal values a won’d of beautyand a
world at peace, and the instrumental values polite and self-controlled, the
standard deviations were somewhat higherthan those for the means of the other
values. This suggests that respondents had somewhat more differing opinions,
hence somewhat less agreement, regarding the relative importance of these
values as compared with the other ones. Yet these differences of opinion were
not significantly greater than others observed. Overall, the responses of the
current sample demonstrated sufficient variability to lend greater confidence to
subsequent analysis.

Given that the current sample was restricted to non-high-school
graduates, it was important to compare their responses to some other sample,

in order to provide a context for more informed analysis. Thus, the responses of

 

143

Rokeach’s non-high-school respondents of 1973 were employed as a way of
explaining and describing the current results.

The parents in the current sample ranked the values on the survey in a
similar way, but not exactly as did Rokeach’s non-high-school respondents in
1973. For example, the current sample was far less concerned with the issue of
war than was Rokeach’s sample. This finding can be attributed to the fact that
the 1973 sample was responding at the time of both the Vietnam War and the
Cold War. The current sample responded years after these ended.

The current sample ranked equality much higher than did Rokeach’s
sample. This may be due in part to Rokeach’s observation from his data that
blacks tended to rank this value more highly than did Caucasians, as the current
sample included proportionally more blacks than Rokeach’s.

Three other terminal values—an exciting life, mature love, and pleasure—
were ranked higher bythe current sample. This indicates that the current sample
was more hedonistic or more concerned with immediate gratification than were
Rokeach’s respondents 20 years ago.

Among the instrumental values, differences between the two groups were
not as pronounced as for the terminal values. Only one value, loving, was
ranked much higher by the current sample than by Rokeach’s sample. Capable
and independent were also ranked more highly by the current sample, but not

markedly so.

144

When compared with the Rokeach sample, the responses of the current
sample exhibited several differences. These differences seem sensible, given
changes in society over time and the differences in racial composition between
the two samples.

The third research question was posed to investigate whether there was
a relationship between parents’ rankings ofthe values and the school attendance
behavior of their children. The question was:

3. Are these rankings by parents related to the school attendance
patterns of their children?

Null Hypothesis 3 stated that there was no relationship between the
parents’ rankings and the children’s days of attendance. This null hypothesis
was rejected for 15 of the 36 values on the survey—eight terminal and seven
instrumental values. Among these 15 values, the strong relationships between
the parental ranking ofthe value and the school attendance of the child were for
the terminal values a comfortable life, a sense ofacoomplishment, and wisdom.
The relationship was negative for a comfortable life, meaning that school
attendance dropped as parents ranked this value more highly. For a sense of
accomplishment and wisdom, the relationship was positive; that is, a higher
ranking of the value was associated with increased school attendance. The
relationships for the other five terminal values and the seven instrumental values,

although significant, were not as strong as forthe above-mentioned three values.

 

145

The second hypothesis was formulated to answer the fourth research
question. This question was:

4. As parents are grouped on the basis of their children’s school
attendance rates, are differences in value rankings evident?

The parents were partitioned into three groups, based on their children’s
school attendance patterns. The partitioning was done to investigate more fully
the relationship between parental value rankings and school attendance, as
confirmed under the first hypothesis. The researcher was now able to compare
a group of parents whose children were highly absent with a separate group of
parents whose children were moderately absent, and then with a group of
parents whose children were not often absent from school. The null hypothesis
was that no differences in rankings would be found.

A number of differences were found in the way parents ranked the values
when they were grouped according totheirchildren’s school attendance patterns.
Underthe first test, using the KruskaI-Wallis, 15 values (seven terminal and eight
instrumental) were ranked differently by the three parent groups. Those 15
differences were then investigated to determine whether the variances in
rankings were due to differences between one pair of groups (between the low-
and medium-absence parent groups, for example), between two pairs of groups,
or among all three combinations of groups.

Use of the Mann-Whitney for this follow-up analysis revealed that three

terminal values were ranked differently in the comparisons among all three

146

groups (a comfortable life, a sense of accomplishment, and wisdom). Eleven
other values four terminal and seven instrumental) were ranked differently in
comparing tow pairs of groups. Two of the terminal values in this second tier, an
exciting life and pleasure, and four of the instrumental values from this set,
ambitious, imaginative, intellectual, and logical, are of particular interest when
considered in concert with the findings under the first hypothesis.

The combined results of the tests for the first and second hypotheses
supported the logicthat certain ranking combinations by parents (value patterns)
might indicate a particular subset of values that could be considered the
"educational values." Future-oriented or delayed-gratification types of values (a
sense of accomplishment wisdom, intellectual) related to school attendance in
a positive way, for as parents ranked these values more highly their children
were more often in school. The converse was also true. As parents ranked
these values as less important, their children were less often in school.

In addition, the parental rankings for such present-oriented or hedonistic
types of values as a comfortable life, an exciting life, and pleasure also related
to their children’s school attendance, but in the opposite direction. As these
values were ranked more highly, the children were less often in school. And
when parents ranked these values lower, the children were more often in school.

Subsequently, the 12-value model of educational predisposition was

formulated in response to the last research question. That question was:

147

5. How accurately does the 12-value model predict the value-ranking
differences found in the study?

The model supported predictions of the direction of parents’ rankings of
values for certain values that were ranked differently by people varying in years
of schooling, based on the results from Rokeach’s national sample.

The model was confirmed in the cases of 8 of the 12 predicted values.
For six of those eight values, the model was particularly accurate. The analysis
employed a combination of the three tests performed under the first two
hypotheses. The strongest relationships were noted between increasing school
attendance and higher parental rankings for the values a sense of
accomplishmentand wisdom. Medium relationships in the same direction were
noted for imaginative, intellectual, and logical. A strong relationship was noted
between decreasing school attendance and higher parental rankings for a
comfortable life.

In addition, there was one other major finding that was not directly related
to the hypotheses, but that did relate to the logic of the study. The families
whose children missed school often were nearly twice as likely to have moved
their residences over the course of the school year, to have had their telephones
disconnected, and/or not to have had telephones, than were families whose
children were usually in school. Families with high-absence children had

considerably more difficulty paying rent and telephone bills than did families with

148
low-absence children, although family income did not vary greatly across the
sample.

This investigation revealed that there were statistically significant
differences in the way parents ranked certain values. Those differences were
associated with their children’s school absence rates. As discussed above, the
findings generally confirm the hypotheses as posed. In addition, the findings
help describe and explain the relationship between parental value patterns and
student absenteeism in a manner consistent with the theoretical background of
the study, as discussed below.

There are parents who ranked more highly such values as a comfortable
life, an exciting life, and pleasure. The same parents ranked less highly such
values as a sense of accomplishment, wisdom, and intellectual. Appliwtion of
the Tonnies (1957) dichotomy precisely predicts these ranking differences.
People who value more highly the present (gemeinschafien) logically value less
highly the future (gesel/schaflen). The findings of this study clearly imply that the
dichotomy exists.

The gemeinschafien social structure emphasizes collectivism and an
orientation to the present. People imbued with this outlook are typified by an
agrarian- or village-style communal life. The alternative structure is that of
gese/Ischafien, emphasizing the furtherance of individual ends. This structure

is common in bureaucratic, urban lifestyles.

149

Schooling in urban settings generally emphasizes the latter structure.
Thus, a potential for conflict exists when a person with the gemeinschafien
orientation interacts with the institution. Brief interpersonal relationships
characterize the institution. What matters to the school isthe future. Conversely,
what matters to the gemeinschafien parent is the present. Forthis parent, family
and community personal relationships, not individual attainment, are the source
of success and joy. The fact that these value-pattern differences were
discovered within a sample that varied little in terms of family income is
significant. The findings indicate that there is not a "culture of poverty," but that
there are differences in value patterns among the poor. Rokeach (1973) made
the same argument, and the current findings are consistent with his.

In addition to examining parental value patterns, this researcher
investigated children’s school attendance. Their days in school varied, and
among students who were not in school often, no verifiable excuses were
offered. In fact, these highly absent, poor-excuse students had their parents’
permission not to attend school. The inevitable interpretation. here was that
parents did not really care about schooling for their children.

It was discovered that the parents whose value patterns were most like
the institutional pattern sent their children to school regularly, given the logic of
the 12-value model and the results of the study. Those parents who had
increasingly divergent (from the mainstream or institutional pattern) outlooks

regarding certain values sent their children to school less and less. Thus, the

150

interpretation is that, as the parent has a greater gesellschaften orientation, the
more likely the child is to interact with the larger society, in this case as a result
of school attendance. The more gemeinschaflen the parent’s orientation, the
less likely the student is to participate in the larger society. These children are
not interacting regularly with the mainstream society. The possibility of social
alienation later in life is more likely under the latter circumstances. The major
finding, however, is that the parental value pattern is, in fact, associated with the
school attendance pattern of the child, in a manner consistent with the Tonnies
model.

Another finding from the study is significant. As shown in Table 3.5 (page
87), there were wide differences among respondents in the number of family
residence changes and telephone "problems." These differences were strongly
associated with absence rates. Family life, under essentially equal financial
support, varied widely within the sample. For children who missed a lot of
school, neither a telephone nor a stable place to stay was certain. High-absence
children were nearly twice as likely as low-absence children to have moved
during the school year. High-absence children were nearly twice as likely to
have had their phones disconnected or get a new phone number during the
school year. And they were more than nine times as likely never to have had a
telephone during the school year.

Children who are absent from school often with no good reason live with

parents who see the world differently. This parental view of the world is different

151

from that held by parents who send their children to school often, or so it was
hypothesized in this study. The difference in world-view can be expressed in
many ways. One measure is a survey of parental opinions, and in this study the
opinions of parents regarding their rankings of values were solicited. Another
measure of world view, however, is direct observation of how a family lives. The
participant-observer method allows for behavioral observation and subsequent
analysis.

The count of residence changes and telephone "problems" forthe sample
in this study provided a behavioral glimpse at family life in the style of a field
study. The count confirmed that differences did exist among families, and it
further confirmed that the differences were aligned with student absence
patterns. These qualitative differences in family behavior strongly suggest that
parents differ in the strength of their commitment to regular school attendance
fortheir children. Moving often and changing the child’s school is, unfortunately,
too often a necessity for many poor people. Yet some move, whereas others
decide to stay to keep the children in the same school. Those who move often
do not place a high priority on school continuity because not only do their
children change schools often, they also miss school often for no "good" reason.
The evidence cited here lends additional support to the other findings that

confirmed the hypotheses of this study.

152

| I' I' [II E' l'

The major implication of these findings is that changing the attendance
patterns of high-absence students in their earliest school years requires
significant attention to parental value systems. Parents whose children are often
absent from school with no good reason see the world quite differently than does
the school or its agents. The anger encountered by truant officers and others
who encourage these parents to send their children to school more regularly is
deeply rooted in cultural differences. Those differences are directly attributable
to differences in value systems between the culture of the parent and the
prevailing culture as represented by the school. Bringing about any significant
change in attendance rates will require significant change in this lack of value
consensus between the home and the school.

From a theoretical perspective, the first observation of note regarding
these findings is that values and behavior appear to be linked. Differences in
parents’ belief or value systems are, in fact, related to their children’s school
absences. The lack of a value consensus between some parents and the school
is associated with higher levels of school absence, whereas a common or shared
value system between other parents and the school results in higher levels of
schoolahendance.

Parsons (1955) wrote, "The conditions under which effective socialization
can take place then will include being placed in a social situation where the more

powerful and responsible persons are themselves integrated in the cultural value

153

system in question“ (p. 17, emphasis added). Children whose parents’ value
systems are not consistent with or integrated with the prevailing social value
system will not be effectively socialized. This lack of socialization is exacerbated
when the value systems are significantly different, and is even worse when the
systems are in conflict.

The problem presented by the poor socialization of some members of
society into the mainstream culture (or into a main variant thereof) is serious from
a pattem-maintenance standpoint. A society that is unable to replicate its
patterns of interaction will not survive long without significant change. The
findings from this study suggest that some poorly socialized members of society,
represented by parents who have value systems that are different from the
mainstream, are essentially preventing their children from attending school. In
this study, it was found that, the more parents differ from the mainstream culture,
as represented by their ranking of values on the survey, the lower their children’s
school attendance. The potential for the production of nonsocialized children is
therefore high, assuming that school is important to the formation of values.

What effect does school attendance have on acquisition of mainstream
cultural values? Parsons (1955) argued that nonfamily institutions have high
levels of influence in highly differentiated societies like that of the United States.
Thus, what level of socialization into the mainstream is evident for children who
are not in school often? Speculation on this question leads to examination of the

problem of school effects.

154

Jencks (1972, 1991 ), Johnson (1985), and many others have argued that
schools have an influence on students. Sometimes that effect is positive; other
times it is not. Often the effects are described in academic terms, less frequently
in social-integration terms. But always, the assumption is that schools make a
difference.

If schools make a positive difference, then the hope of society’s wishing
for some stability and replication is that children are in school, where they will be
socialized. If the school influence is in a negative direction, the hope of society
for replication lies in the parents' having an integrative effect on the child. But the
implication of this study is that, among the poor, children who miss the most
school live in families that are least like the mainstream society. The
socialization of these highly absent children is in the hands of families that see
the world differently from the prevailing norms. The chances that such families
will support their children’s positive integration into mainstream society are
indeed slim.

A possible solution to this problem is to secure a greater value consensus
between the family and the school. Herein lie the practical implications of this
study. There are a number of current and past practices, and some suggested
future actions, that aim directly at the value consensus problem. The primary
solution is first to establish face-to-face, personal relations with parents of high-
absence students. Then, educators can concentrate on working with such

parents to change their value patterns.

155

High, unexcused student absenteeism is not a problem in all schools for
a large proportion of the student body. It is a problem, however, for any
individual child for whom the pattern is a reality. Thus, for at least the one case
that may exist in a school, or for the many, educators can and should adopt
practices that initiate and support fundamental home-school cooperation. The
vaunted mission ofteaching and learning is meaningless for absent children, and
educators need not apologize for substituting some alternate goals. Finding
ways to bring children to school on a daily basis is a prerequisite to instilling a
love of learning, an ability to read, or any of a number of other worthy aims. A
simple reorientation of philosophy, with a minor redirection of existing resources,
is highly appropriate.

In the one-room school of old, a teacher taught an individual student for
many consecutive years. The relationship among the student, the student’s
family, and the teacher was generally long term, severed only when the teacher
or the family moved away. There was not a lot of moving away then; it was a
time when children were often born in the same house in which their parents
were born. This pattern of multiple interactions overtime between the family and
the school supported a higher level of value consensus, due in part to the on-
going nature of the social relationships. Current "restructuring" proposals in
elementary schools (and middle and high schools as well) include teachers
having the same group of students for two or more years. Although other

pedagogical benefits are generally advanced in support of these proposals, the

156
opportunity to bring into alignment the school and home value patterns is obvious
and is strongly supported by the current research.

Another recent suggestion is called “invitational education,“ but it again
reflects behavior long used by excellent teachers. Advocates of this practice
suggest altering educational operations to make them more interesting, less
harsh, more open, and therefore more "inviting." Negative comments in red pen
on the front of a student’s paper ("NOT his best effort!“) carry a strong message
in word and in medium. A positive message, in blue ink, on the bottom or back
of the paper is more invitational ("I’d love to see Johnny able to put on paper
what I’m sure is in his head"). The parent who already has a negative
orientation toward institutional life will find a decreasing use for schooling when
the experience appears to berate his or her child. The parent in this instance is
much less likely to interact with the school. The opportunity is thus greatly
diminished to reach the parent in order to secure a change in his or her values.

A similar case can be made for the difference between sending home a
typed note on letterhead as opposed to a handwritten note from either the
principal or the teacher. There are many other examples. The challenge here
is to think in terms of what might “soften" the harshness of the admittedly
necessary rules and regulations inherent in a larger bureaucracy in orderto have
a better chance of welcoming certain families into the larger community. It is not

required to abandon or break rules, but merely to administer them in a less

157
aggressive or assertive manner, or in a manner that appears more humane and
understanding.

Another arena of action is that of interagency cooperation. Certain
demographic changes over time have presented schools with seemingly
insurmountable obstacles. Writing a nice note home seems to be a very weak
practical suggestion when the parent who receives it has characteristics such as
these: didn’t finish high school, functionally illiterate, 19 years old, four children
(the oldest of whom is in kindergarten), no husband, abusive boyfriend, no job,
and no other family support or prospects. This scenario is altogether too
common. Schools have sought to connect with other community agencies (both
governmental and private) to help provide food, shelter, clothing, and other
support to such families. It is known that schools alone do not have the
resources to solve such difficult problems.

Yet in pursuing cooperative ventures, educators could insist on developing
common goals that the current research suggests as critical. For example,
keeping families in the same house for several years’ time would certainly
increase the chances that teachers could establish longer-term relationships with
parents. Suchagoal is compatible with the missions of other agencies. The fact
that it also affords an opportunity to the school to secure a higher values
consensus with the family is an additional benefit.

This focus on changing the marginal parents’ values system is critical to

the delivery of educational services to their children. Many people understand

158

and accept that entering kindergartners have a lot to learn about getting along
with others, particularly in a large group setting. This is an appropriate model for
practical application of the findings of the current study to adults. We should be
careful to preserve and, in fact, to celebrate those individual and cultural
differences that provide a richness of diversity within society. We have an
obligation as public educators to accept children of all backgrounds. But we also
have a mission to provide for them a foundation of opportunity, so that as they
progress through our institution they increase in their ability to function in our
complex society. One method strongly suggested by this study is to identify the
children who are very often absent and to target their parents, to bring both the
child and the parent into the larger society, and to introduce both to the notions
of getting along in the larger world.

From using the island tradition of oral storytelling to help teach reading in
Hawaii, to providing a hot lunch in Maine, public schools must continue to reach
out to the immediate community, become a part of it, and use that relationship
to build bridges for families in desperate need.

The efforts of educators to understand and accept varying value patterns
and alternative family life circumstances will serve to shrink the cultural gap
separating some children from access to the larger world. Because of the highly
complex nature of both the systems of social interaction and the personal value
patterns held by members of society, no single program or emphasis is a

panacea. Rather, skilled educators will solve the problem of some poor

159

children’s missing a lot of school for no good reason one family at a time, using

a variety of techniques.

BecommendaticnfloLEunthesearch

The results of this study provide a background for further research in four
major areas. The nature of value patterns among parents, patterns of school
absence among their young children, the relationship between the two, and the
possible existence of a set of “educational values“ all merit further investigation.
It is suggested that this study be replicated with larger samples, nonurban
samples, and more affluent samples.

How consistent are differences in parents’ value rankings? In this study,
certain trends were found, particularly with regard to six values. Do these same
patterns exist in other samples? For example, are there affluent parents who
favor immediate gratification over saving for the future? Conventional wisdom
would answer in the affirmative, but updating of Rokeach’s national sample
would be of benefit.

Regarding the nature of school absence, school administrators and board
of education members commonly accept the notion that an attendance rate of
90% or better is tolerable, and perhaps it is. But this study and other inquiries
into school absence have suggested that the nonattending population includes
a significant number of children who frequently miss a lot of school. The

common perception is that the 10% of students who are absent on a given day

160

evidence random absence patterns: a few sick, a few doctor’s appointments,
and so on. In fact, the absentee population has a significant membership whose
absence pattern is very regular and whose excuses are weak, at best. More
information, and particularly updated information, would be very helpful as
American educators are being asked to consider lengthening the school day and
yean

In this study, a relationship was found between parents’ rankings of
certain values on the Rokeach Value Survey and their children’s school-absence
patterns. Does the same relationship exist in samples that are not poor? Is there
a difference between urban and rural, rural and suburban, or other stratified
samples? Answers to these questions would provide a better perspective on the
importance of the role played by value-pattern variance in explaining school-
absence patterns.

Another important area of inquiry concerns the question of whether a set
of educational values exists. A logical proxy for educational values was tested
in this study, and the results were mixed. Six values related fairly strongly to
school attendance, but no simple explanation for all of the results of this study
was supported. Testing the 12-value model posited herein on other samples is
suggested. This investigation was limited to families in which the mother
reported having no high school diploma. Would the results be similar in samples
of parents having a range of educational attainment? Can the value mix be

linked to other school-supporting behaviors, such as voting on school millages

161

and board elections, volunteering in school, donating money, and the like? Also,
pattern analysis of larger data sets might be helpful in determining whether an
educational-values prototype exists.

Finally, if other research supports the idea introduced in the current study
--that the values of the parent are controlling in the life of the child--it is sensible
to design and test some intervention programs. Schools will continue to exist,
and parents will continue to keep their children at home against strong social,
legal, economic, and perhaps moral sanctions. Less than 6% of the population
in this study exhibited the high-absence, poor-excuse behavior, so the number
of cases was not large. But the possible effect on the larger society is large, so

it may be worthwhile to study effective responses to this on-going problem.

mm
The fact that a few children account for a large percentage of school
absence is bad news. The fact that many of these students have poor excuses
(if any) for missing school ought to alarm more educators than it does. The
knowledge gained in this study--that a relationship exists between school-
absence patterns and parent value patterns-~is not surprising, nor is it particularly
newsworthy. However, the implications of the logic of this study are significant.
The postindustrial society the United States now embraces has resulted
in a marked decrease in manual-labor opportunities, coupled with a dramatic rise

in a service-oriented economy. The need to read, write, and get along with

162

others is significantly greater now than in even the recent past. True, many
places of business use technology to compensate for deficiencies in their
employees, as pictures of products on the cash register keypad instead of
numbers for prices would indicate. But the level of skill necessary to function in
this complex society continues to rise. Regardless of variations in definitions of
success, the hermit and the street dweller who happily anticipate a todayjust like
yesterday are rare.

The one common opportunity in support of having a reasonable set of
options for life as an adult is schooling. The knowledge and skills, and
particularlythe social integration, availablethrough universal public education are
crucial to the development of the individual. They are also paramount to
cohesiveness and order in a highly differentiated society. Considerjust one facet
of many, that of society’s general acceptance of a wide variety of cultural and
linguistic differences. This is possible only through the transmission of the
valuing of that diversity to a critical mass of the society. Should American society
be prevented from modeling and nurturing these notions with the young, the
potential for cultural, ethnic, and racial conflict will escalate. Children who miss
school are likely never to encounter, much less accept, people who are not like
themselves. Social disintegration, in the true sense of lack of integration, is
inevitable.

The problem is not so much the question of how large a nonsocialized

segment of society can be accepted before the nation gets in serious trouble.

163

Rather, the problem is how many cases should be accepted. As a concomitant
to any public or personal oath of educational purpose, members of society ought
to commit themselves to bringing all children to the common experience of the
society-school. In this regard, through the support of public education, people

are in a true sense promoting the common good.

APPENDICES

APPENDIX A

ROKEACH’S TERMINAL AND INSTRUMENTAL VALUE MEDIANS

164

389.5 3:... 3.39:8 .385533. c. .2... must—:2 5.52: 0.... :39? Scam.“—

 

 

 

l 3:3... 3 .2 3 .2. 3.. N3 3: 2.. 3:2: 3 .3... 3.2.89.3
.8. 3 :3 3 .43 3 .33 3 . .3 3 .3... 3 .3... 3 .3... o.......§.3...
.8. E. 5.. 3.. .3. 3:3... 3: 3.... 3.. 2.. E. 2.. 3: 3.... 2.3..
.8. 3:33. 3.. 2... 3.. 5.. 3.. .3. 3: 3.: 3: S. 3.. 2. .5380
l 3:2 3:38. 3:33 E. 33 8:3... 3.. 2.. 3 .8... 358..
.8. 3.. 3... E. 3.2 3.. .3. E. 3.3. E. 3.3. E. E. E. «.2 .8....3
.8. 2...; 3 .3. 3: 3.... 3.. .3. 3:2. 3:3. 3:32 328:2...
l 3 .2. 3 .2. 2:3. 3:3. 3:3... 3:3. 3:3... .5888...
.8. 3.. ca 3.. 3... E. 3.. E. 2. 3.. 3.3. 3.. 3.3. 3.. 3.2 2.2.532...
I A. .33 c .43 A. .33 2 ....3 c .33 c .33 c .33 .88:
.8. 3 .3... 3 .3... 3 .3 3 .3. 3 .2 3 .2. 3 .3... .35.:
.8. 3 .3... 3:2: 3 .3... 3 .2 3 .3 3 .33 3 .33 9.29....
.o. 3 .8... 3 .3 3 .2. 3 .3... 3 .2 3 .33 3...... 389.58
.8. 3:3. 3:33. 3:3... 3 .3... 3 .3... 3 .33 3 .2 :36
.8. 3:3... 3:3... 3:3... 3:3. 3 .3. 3 :3 3 .2. 3.32.0
l 3 .2. 3:2. 3 .2 3 .83 2.5.... 3:3,... 2:33 “.335
l 3 .33 3 .33 3 .2 3 .3... 3 ..... 3 .3. 3.:3 88.268...
l 3 .o... 3 .2 3 .E 3 .33 3 :3 3 :3 3 .3 32.35...
a 3. 8 82 3% 3.3 33 .3 u 2 is.
.33. .35... .3... .55... .30 2.8.3. .5. $3.: .83 $3.: .3 3-3 .3 Te
.3235 2.6%

 

Avon... R Z. ZO:.<UDQm

Z— OZ_>¢<> meOMO d0... mzmfldo ¥Z<¢ which—ECU DZ< mZ<_Qm2 mDA<> A<._.Zm_23~:.mz_ .H< Mal—mgr

165

3.3.8 :5: 2.3268 $635222. 2. .22.. 3:32.... 5:58 8.. 239.... 352....

 

 

.8. 3 .3... 3 .3 2. :3 3 .2 3 .3. 2:3 2...... 83.3
.o. 3 .3 3 .2 3:2: 2: 3. 3:2 3 .2 3 .3 2.85... 2.;
8. 3.. 2.. E. 2. E. .3. 3: 3.... 3.. 2.. E. 2. 3: 3. 8:388. 3.83
.o. 3 .3... 3 .33 3 .3 3 .3... 3 .2 3 .2. 3 .3. 383......“
.8. E. .3. 3:3. 2:2: 3 .3... 3 .2 2. .2 3 .3 8.328
.8. 3.. 3.. 3.. 3. 3.. 3.... E. 3.... E. 3.... 3.. 2. 3.. 2. 2:23...
.o. E. 2. 3.. o... 3.. 8. 3.. 3 3.. a... 3.. 3 3.. 3. 3.3.... 38.32
.8. 2.. 8. 2.. 3. 2.: 2. 3.. 2. 2.: 2. 3: 2. 3:2. 26. 8...:
.8. 3 .3 3 .3 3 .2 E. 2: 3...... 3...... 3 .2: .88.... :8.
.8.. 3:2 3:2: 3 .2 3 .2 2. .2 3 .2 3 .2 3.....8:
l 3 .2 3 .2 3 .3 3 .2 3 .2 3 ..... 2. .8... .83...
.8. 3 .3 2 .3 2 .3 2 .2 3 .2 3 .3... 3 .2. 2.5.3 .8....
I 3 .3. 3 .2 3 .3 3 .3 3 .3 3 .3... 3...... 2.38..
l E. m... 3.. 2. 3.. 3. 3: 3.. 3.. 3. 3: 2. 3.. 3. 222 .o .8.: <
.8. 2 .3 3 .3. 3 .2 3 .2 2 .2 2 .3 2 .2 83,. ... 2.3, <
.8. 2. .3 3 .3 3 .32 3 . 2. 2: 2. 3.. .2 .2. 3. .._2._..,....:_3.....c 2...; <
I 2... 2. 3.. 3. 3.. 3. 3:3. 3:3. 3: 3. 3:3,... 9.... 8...? ._<
.8. 3:2. E. 2. 3...... 3:3 3 .8. 3 .2 3 .3 a... 22......53 <
.. .3 8 8. 3% s... a... x. u 2 2......
.33 .8.... :6 58.. .8.. 28,3. .3. 3...... .23 3...... 5. 3-3 s. I.
diob 2.5.0

 

 

3.3...— H 7.. 7.:— _ (.932...

7: OZ.>¢<> $230.20 go“. mszzC ¥Z<z MEmOL—AOU CZ< mZ<_Cm2 m_3-_<> ASL—21.2. . N< ,.,_._:<._.

APPENDIX B

THE ROKEACH VALUE SURVEY

166

ROKEACH
VALUE
SURVEY

Milton Rokeoch

 

Nome:
Date of Birth:
City & State of Birth:

 

 

 

Sex:

 

0 Consulting Psychologists Press, Inc.
3803 E. Boyshore Rood

Polo Alto, CA 94303
0725

167

Instructions

The following page lists 18 values arranged in alphabetical order. Each value
is accompanied by a short description and is printed on a gummed labelthat
can be peeled off easily and placed in the boxes in the left-hand column of
the page.

Your goal will be to rank each value in its order of importance to you. Study
the list and think of how much each value may act as a guiding principle in
your life.

To begin, select the value that is of most importance to you. Peel off the
corresponding label and place it in Box 1. Next, choose the value that is
second in importance to you and place its label in Box 2. Work your way
through the list until you have ranked all 18 values on this page. The value that
is of least importance to you should appear in Box 18.

When you have finished ranking all 18 values.turn the page and rankthe next
18 values in the same way.

When ranking, take your time and think carefully. The labels can be moved
from place to place so you can change their order should you have second
thoughts about any of your answers. When you have completed the ranking
of both sets of values. the result should represent an accurate picture of how
you really feel about what’s important in your life.

 

Rokeach Value Survey—Form C. Copyright © 1983 by Milton Rokeach. Published in 1988 by
Consulting Psychologists Press, Inc. 3803 E. Bayshore Road, Palo Alto, CA 94303. All rights
reserved. No part of this booklet maybe reproduced by any means without written permission
of the publisher. Printed in the USA.

/

 

 

lfilléllallfillallEllilla‘llmllmllulmlmllsllwllmll-‘l

 

_3
CD

 

 

1

O)

8

 

 

 

ll3 la‘llcol co lullmllwllsllwllwllal

ll-‘ll-‘ll-‘l'tll-‘llt
_\lau_ai|.l>floo,r\>

_\
00

 

at

%**

A COMFORTABLE LIFE
a prosperous life

EQUALITY
brotherhood and equal opportunity for all

AN EXCITING LIFE
a stimulating. active life

FAMILY SECURITY
taking care of loved ones

FREEDOM
independence and tree choice

HEALTH
physical and mental well-being

INNER HARMONY
freedom from inner conflict

MATURE LOVE
sexual and spiritual intimacy

NATIONAL SECURITY
protection from attack

PLEASURE
an enjoyable. lelsurelv lite

SALVATION
saved: eternal life

SELF-RESPECT
self-esteem

A SENSE OF ACCOMPLISHMENT
a lasting contribution

SOCIAL RECOGNITION
respect and admiration

TRUE FRIENDSHIP
close companionship

WISDOM
a mature understanding of life

A WORLD AT PEACE
a world free of war and conflict

A WORLD OF BEAUTY
beauty of nature and the arts

’1\ l I

When you have finished. please proceed
to the next page.

 

 

 

tallillallallfllallsllilESlcolloollJimllmllsllwllmllal

 

 

 

laslla‘llallall:llallsllillallmllmllsllmllmllsllwllmla

Please rank these values In the some manner
I as you did on the preceding page.

AMBI'I'IOUS
hardworking and aspiring

BROAD-MINDED
open-minded

CAPABLE
competent: effective

CLEAN
neat and tidy

COURAGEOUS
standing up for your beliefs

FORGIVING
willing to pardon others

HELPFUL
working for the welfare of others

HONEST
sincere and truthful

IMAGINATIVE
daring and creative

INDEPENDENT
self-reliant: self-sufficient

INTELLECTUAL
Intelligent and reflec tive

LOGICAL
consistent; rational

LOVING

.1 affectionate and tender

LOYAL
faithful to friends or the grow

OBEDIEN'I‘
dutiful: respectful

POLITE
courteous and well-mannered

RESPONSIBLE
dependable and reliable

SELF-CONTROLLED
restrained: self-disciplined

S s s%
’41 ’I l\ ’m
7m: 711% ’1’“?

APPENDIX C

COVER LETTER

170

Dear Parents,

I’m an elementary principal in Lansing and a graduate student at Michigan State University.
As part of my studies I’d like your help through filling out the attached questionnaire.

I have a very high interest in knowing more about family values. About twenty years ago
there was a lot of research about values, but it is now out of date. My study is intended in part to
bring up to date our knowledge about this important area.

What I hope to learn through your help is how schools and families can work together more
closely to ensure that students have solid academic progress. Knowing more about family values will
help schools do a better job with students.

I think you will enjoy working through the Values Survey. It will take about a half an hour.
Notice that you will be able to change your answers by rearranging the stickers, and of course there

are no right or wrong answers or hidden meanings. I chose this survey because it is straightforward
and actually fun to do.

I would like to have the parent who has the most "day to day" contact with your child or
children to complete the survey. This way, the connection between the values of the parent and the
child is most strongly represented. You were selected on a random basis from among all the parents
in the Lansing School District.

Your name and responses will be kept strictly confidential. The number I have put on your
form is only there to help me keep track of the forms as they come in. Once I record your responses
this number will be cut off from the survey, so you cannot be identified. I have also crossed at? the
name portion on the survey.

Your participation is strictly voluntary and there are no consequences should you choose not
to complete the survey. You might fill out the survey, mail it, and later decide you don’t want to
participate. Simply give me a call and you will be excluded, no questions asked.

By completing this form you are voluntarily agreeing to participate in this study. If you would
rather not, just write "no thanks" on the instruction page and mail the survey back in the envelope
provided. .

If you have any questions, or would like to know about my study when it is finished, feel free
to call me at Lewton Elementary School, 374-4393, during school hours.

Thank you very much for your help.

Sincerely,

é+qi~i2tluL

Stephen R. Hecker

BIBLIOGRAPHY

BIBLIOGRAPHY

Adams, A. V. (1978). Ibeingeunganslsoiyamtmnemoloyment
Kalamazoo, MI: W. E. Upjohn institute for Employment Research.

Allport, G. W., &Odbert, H. S. (1936). Trait-names: Apsycho-lexical study.
BayQIJQIQgiQaLMQDQQLaphs. .41 (Whole No. 211).

Anderson, E. (1991). Neighborhood effects on teenage pregnancy. In C.

Jencks 8. P. E. Peterson (Eds), Inaumanyndamlass (pp. 375-398).
Washington, DC: The Brookings Institution.

Babble, E. R. (1973). W. Belmont, CA: Wadsworth.

Beech, R. P. (1967). Value systems, attitudes, and interpersonal attraction.
DissertatiQnAbstractantematicnaL 28 1125A.

Berg, 1. (1980). Absence from school and the law. In L. Hersov &l. Berg

(Eds..) WWW
Leiusai (pp. 137- -.148) New York: John Wiley & Sons.

Berg,l. ,,Brown 1., &Hullin, R. (1980). W

a... l-.I I Qq'l. ‘ I0. el‘ewll

New York: Springer—Venag.

Brown, M. A. (1976). Values-—a necessary but neglected ingredient of
motivation on the job. AcademLManagemenLBeyiew 1(4).15-23

Boudon, R. (1981). WWW
analysis. London: Routledge & Kegan Paul.

Buenning, M., 8. Tollefson, N. (1987). The cultural gap hypothesis as an
explanation for the achievement patterns of Mexican-American
students. W. 24(3), 264-272.

Buros, O. K. (1978). Ibuighhmentalmeasurementyeamookwol. I, pp.
1028-1033). Highland Park, NJ: Gryphon Press.

171

172

Bury, l. B. (1970). A study of the effects of air pollution on children. JaumaLoj
Schmuzieaiih. 4.0. 510-513.

Butler, D. C. (1973). An analysis of the values and value systems reported by
students, the general public, and educators in a selected Appalachian
public school district (Doctoral dissertation, Michigan State University,
1973). QisseriatiorLAbstEactsJMemationaL 14. 6015.

Central Advisory Council for Education (CACE). (1967). W
primazyscmnls (The Plowden Report). London: HMSO.

Cervantes, L. F. (1965). WWW. Ann Arbor:
University of Michigan Press.

Cherry, N. (1976). Persistent job changing—ls ita problem? dogmaLoj
OccupationaLEsyohology. 4.9, 203-221.

Chilcott, J. H. (1985). Yaqw world view and the school: Conflict and
accommodation. JoumaiaLAmencaandIaniducatlon. 25(1), 21-32.

Children’ 5 Defense Fund of the Washington Research Project, Inc. (1974).
SEW. Washington, DC. Author.

Clark, J. L. (1983). Maluesandacademiaachieyementamonamraiindian

W. Grand Forks: University of North
Dakota.

Clark, R. (1983). EarnilyliteandschoaLachieyememLflnyoooLblack
W. Chicago: University of Chicago Press.

Clarke, R. (1980). Absconding from residential institutions for young
offenders. In L. Hersov&l. Berg (Eds..) QIII.Qf.§QIJQ_QIJ_M9_deEn

WW (pp. 111- -136). New York:
John Wiley & Sons.

Coldiron, J. R, &Skiffingion, E. W. (1975). Condition variables in relation to
measured student performance: The use of indicators' In Pennsylvania.

Cooperative Accountability Project. In SEW
(Educational Research Service, 1977).

Courtney, R. (1986). islands of remorse: Amerindian education in the

contemporary world. CumisqumJnguinr, 16(1), 43-64.

173

Crane, J. (1991). Effects of neighborhoods on dropping out of school and
teenage childbearing. In C. Jencks 8 P. E. Peterson (Eds), Ih.e_uEb_an
undamlass (pp. 299-320). Washington, DC: The Brookings Institution.

Cuch, F. S. (1987). Cultural perspectives on Indian education: A comparative

analysis of the Ute and Anglo cultures. Egyjtyandfixcaliancs, 23(1-2),
65-76.

Dawson, G. E. (1896). Astudy in youthful degeneracy. Eadagogjsal
Seminary. 4(2). 221-258.

Dillon. HJ- (1949). EanyscthieamAmanEeducatronaLnrnblem. New
York: National Child Labor Committee.

Douglas, J. W. B., 8. Ross, J. M. (1965 February). The effects of absence on

primary school performance. BubshMmaLoLEducatnnalfisychology.
pp. 2840.

Duncan, G. J., & Hoffman, S. D. (1991). Teenage underclass behavior and
subsequent poverty: Have the rules changed? In C. Jencks 8. P. E.

Peterson (Eds), Inaumanundamlass (pp. 155-174). Washington, DC:
The Brookings Institution.

Durkheim, E. (1951). W. Glencoe, IL: The Free

Press.

Eaton, M. J., 8. Houghton, D. M. (1974). The attitudes of persistent teenage
absentees and regular attenders towards school and home. Jn'sn

JoumaLoLEsychology. 2. 159-175.

Educational Research Service. (1977). W. Arlington, VA:
Author.

Eswara Reddy, V. (1977). Iheoumtschoglmn. New Delhi: Sterling

Publishers.

Farrington, D. (1980). Truancy, delinquency, the home, and the school. In L.
Hersov &l. Berg (Eds..)
andssthLLeIusapr. 49—64). New York: John Wiley&Sons.

Feather, N. T, & Cross, D. G. (1975). Value systems and delinquency.
Parental and generational discrepancies in value systems for delinquent

and nondelinquent boys BmIanmIEnamLSooIaiflinIcaLEsycholngy.
1.4., 117-129.

174

Ferguson, T. (1952). WWW. London:
Oxford University Press.

Fogelman, K. (Ed.). (1976). W London: National

Children’s Bureau.

Fogelman, K. (1978). School attendance, attainment, and behavior. British
JournaLoLEducationalEsycnology.A8.148-178.

Freeman, R. B. (1991). Employment and earnings of disadvantaged young
men in a labor shortage economy. In C. Jencks 8 Peterson (Eds.), 1113
umanyndamlass (pp. 103-121). Washington, DC: The Brookings
Institution.

Frerichs, A. H. (1969). Relationship of elementary school absence to
psychosomatic illness. JaumaIELSQIJmIflealih. 38. 92-95.

Galloway, D. M. (1980). Problems of assessment and management of
persistent absenteeism from school. In L. Hersov8l. Berg (Eds), Qut

WW (pp 149-
170). New York: John Wiley 8 Sons.

Galloway, D. (1976). Size of school, socio-economic hardship, suspension
rates and persistent unjustified absence from school. BmIangnmajmf

Educationalfisychology. 46. 40-47.

Galloway, D. (1982). A study of persistent absentees and their families.
BntisndoumainLEducatIonaLEsycholoay. 52. 317- 330.

Galloway, D. (1985). MW. Elmsford, NY:
Pergamon Press.

Gilbert, 8. E. II, 8 Gay, G. (1985). Improving the success in school of poor
black children. EIJLDEIEJEQIED. 51(2). 133—137.

Goss v. Lopez, 419 US. 565 (1975).

Gray, G., Smith, A., 8Rutter, M. (1980). School attendance and the f' rst year
of employment. In L. Hersov8l. Berg (Eds.,) Qut_Qf_sch9_Qi;_M_Qd_e_[n

WWW (pp. 343- -.370) New York:
John Wiley 8 Sons.

Greenstein, T., 8Bennett, R. R. (1974). Order effects In Rokeach’svalue
survey. JouLnaLoLBeseaLcthEeLsnnality. 8. 393-396.

175

Hague, w. J. (1968). MaluesystemsandxocatmnaLchorceoLthapnestnood.
Unpublished doctoral dissertation, University of Alberta, Edmonton,
Alberta, Canada.

Hansard. (1974). Parliamentary Debates. flouscctgommonscfficiaueoon
(Fifth Series, Vol. 877). London: HMSO.

Healy, W. (1915). Ihojnojyiouadelinoucnt. London: Heinemann.
Hecker S. E. (1953). EaflLSQhQQIWKEDIUQKLAfludLSQQDSQLQd

I‘ A II A n ' ._ 'I 0 0 v1. ‘ .I0- I 1'..."

Schools. Lexington: University of Kentucky.

Henderson, D. W. (1973). A measurement and analysis of value systems and
behavioral change as related to a public executive development

workshop. DissertationAbstractanternationaL 8.4.. 3514A.

Henneghan, K. (1971). Won!

WWW. Unpublished doctoral
dissertation, Michigan State University.

Hersov, L. A. (1960). Persistent non-attendance at school. qumaLQLthid
Esychnlogyandfisychiatry. 1. 130-136.

Hersov, L., 8Berg,l. (Eds..) (1980).Qu1_QLSQh9_QL_M.Qd8En_p_QEspe_cinLe§_In
tmanmLanthQLEeIusal. New York. John Wiley8Sons.

Hollingshead, A. B. (1949). Eimjownsyouth. New York: John Wiley 8 Sons.

Hopkins, K. D., &Glass, G.V. (1978). Basicsiaflsflcsjorjhabehayiorai
sciences. Englewood Cliffs, NJ: Prentice-Hall.

ISTD. (1974). Truancy in Glasgow. EmianouLnaLoLCflminolooy 1.4. 248-
255.

James, T. (1988). Rhetoric and resistance: Social science and community

schools for Navajos In the 1930’ s. WED—OHM!-
28(4), 599-626.

Jencks, C. (1991). Is the American underclass growing? In C. Jencks 8 P. E.
Peterson (Eds.), Ibumanunderclass (pp. 28402) Washington. DC:
The Brookings Institution.

Jencks, C. (1972). incouaiity. New York: Harper8Row.

176

Jenkins, C. G. (1974). Value differences in adolescent drug abusers and non-
drug abusers (Doctoral dissertation, Arizona State University, 1974).
DissenatiorLAbstractanteEnationaL Z5 10, 044.

Johnson, N. B. (1985). Westhaven. Chapel Hill: University of North
Carolina Press.

Jones, A. (1980). The school” s view of persistent non-attendance. In L.
Hersov 8|. Berg (Eds.,) '

andehQQLEeIusapr. 171- 1.88) New York: John Wiley8Sons.

Kahn, J. H., 8 Nursten, J. P. (1962). School refusal: A comprehensive view
of school phobia and other failures of school attendance. American
JoumaLotQEthoosichiatEx. 82. 707-718.

Kitwood, T. M., 8 Smithers, A. G. (1975). Measurement of human values: An

appraisal of the work of Milton Rokeach. EducationaLRcsaarch
(England), 11(3), 175-179.

Klein, E. (1945). The reluctance to go to school. IhaEsychoananojtocyoj
th.e_Qh]!d. 1. 263-279.

Kline, L. W. (1897). Truancy as related to the migratory instinct. Eadagogicai
Seminary. 5. 381-385.

Lummis, C. (1946). The relation of school attendance to employment records,

army conduct, and performance in tests. BritishJaumaLaLEducaflQna!
Esychology.1_6,13-19.

Macmillan, R. W. (1968). Astudmfihaefieflctsocioeconommmrscn

I‘ In -0:I‘ ‘ll‘l 0 0-.I row.“ In P‘QII‘
International Reading Association. (ERIC Document Reproduction
Service No. ED 026 217)

Mahoney, J. (1974). Seif—actualization and value stmcture: An empirical

investigation. Esychologyficnons. 35(2), 979-985.

Mahoney, J. (1977). Values and neurosis: A comparison of American and

Israeli college students. WW. 192(2). 311-312

Mahoney, J., 8 Katz, G. M. (1976). Value structures and orientations to social
institutions. W991. 9.3. 203-211

177

Marascuilo, L. A. 8McSweeney, M. (1977). Noonaiameincandcjstnhuflon;
freemeihodsioflheeoeiaieeienees MonterGY. CA: Brooks/Cole

May, D. (1975). Truancy, school absenteeism, and delinquency. Scottish
EducationaLStudies. Z. 97-107.

McCarthy, J. J. (1972). Congruence of family value systems related to
unobtrusive measures of their behavior (Doctoral dissertation,

University of Notre Dame. 1972). mssenmionAbstractstemationai.
22. 33,324.

McClelland, D. D. (1970). MaluesJaluesystemsEanthejeiLeiopmentai

sinietureetmeraljudgment. Unpublished masters thesIs Michigan
State University.

McMillan, S. I-I., Jr., 8Behrman, C. A. (1987). Dropping out: QUIet kIlIer of

the American dream. In QuflreubledJeensuSen’ousnroblemsare

cdpoflntheJiyeseLmflflenseLAmeneasyounepeople. Pittsburgh:
Public Television Outreach Alliance.

Mendelberg, H. E. (1986). Identity conflict in Mexican-American adolescents.
Adolescence 21(81). 215-224.

Moore, T. (1966). Difficulties of the ordinary child in adjusting to primary
school. JenmaLoiChfldfisyobeloeLandfisycniatnr. Z. 17- 38.

Mueller, D. J. (1974). A test of the validity of two scales on Rokeach’s value
survey. JeumaLeLSoeiaLstchelegx. 911(2), 289-290.

Mumfmdv L- (1925)- IbegeldendasttudnnAmencanexpenenceand
minim. New York: Boni 8 Liveright.

Natriello, G., McDiII, E. L., 8 Patios, A. M. (1985). School reform and potential
drooouts. EdueaflonaLLeadeEsnip. 43(1). “44

Newson.J.. &Newson. E. (1977). Eerspectileseneehoelatsexenyeamld.
London: Allen 8 Unwin.

Ogbu J. U. (1978). MinemyedueatienandeasteJheAmeneaneystemJn
crosscuitiiraioaLsocctiye. New York: Academic Press.

178

Ogbu, J. U. (1988). Diversity and equity in public education: Community
forces and minority school adjustment and performance. In R. Haskins
8 D. McRae(Eds.),E.Qfl918§_LQLAmEHQaLS_pubfi£_S.QhQQI§:_Ieth8Es.
W(pp.127-.170) Norwood, NJ. Ablex Publishing.

Okey,T. N. (1990). IheJamIleeEsmmethenndmduaLLdeeisienjedrep
mammal. Unpublished doctoral dissertation, Michigan State
University.

Pars°nsv T., 8‘33'93. R- F- (1955). W120.
process. New York: The Free Press.

Parsons,T.,8ShiIs, E. (Eds). (1951). Iowamecenerauneonretadien.
Cambridge, MA: Harvard University Press.

Parsons, T., Shils, E., Naegele, K. D., 8Pitts, J. R. (Eds). (1961). Ihoofiascf

society._Eoundatiens_of_medem_seeielegieal.tnought New York: The
Free Press.

Reid, K. (1985). Imancyandeeheolebsenteeism. London: Hodder8
Stoughton.

Reynolds, D. Jones, D., St. Leger, S. 8 Murgatroyd, S. (1980). School
factors and truancy In L. Hersov8i. Berg (Eds) Qin_Qf_schoc|;_

Modemperspectiyesensenoeuetusal (pp. 85-110). New York. John
Wiley 8 Sons.

Riesman, D. (1952). Eacesinjhecrowd. New Haven, CN: Yale University
Press.

Riessman, F. (1962). W. New York: Harper&

Row.

Robins, L. N., 8 Ratcliff, K. S. (1980). The long-term outcome of truancy In

L. Hersov8i. Berg (Eds). W11
tnranmndeeheolleiusaupp. 65-84). New York: John Wiley8Sons.

Robinson, P. E. (1977). Value patterns, social inheritance, and educational
systems. DissertatienAbstEaetsinteEnatIonai. 31. 6775A.

ROKGaCh» M- (1968). WWW
anchhange. San Francisco: Jossey—Bass.

179

Rokeach, M. (1973). Ibenaturecthumanxalues. New York: The Free
Press.

Rokeach, M. (1980). Some unresolved issues in theories of beliefs, attitudes,

and values. In H. E. Howe8M. M. Page (Eds) Neblzaskafixmpesium
on_Mo_tiyation_1_919(pp. 261- -.304) Lincoln: University of Nebraska
Press.

Rozelle, R. M. (1968). The relationship between absenteeism and grades.
EducationalandEsychelegicaLMeasuEement. 281151-1158

Scottish Education Department. (1977). IEuancLandJndiseipline (The Pack
Report). London: HMSO.

Siegel, S. (1956). NenparametflcstatisticsfcuhebehayieraLsciences. New
York: McGraw-Hili.

Sinclair, R. L., 8Ghory, w. J. (1987). Reachinomarginaletudentsaarznmarv
concechLscnchEnewaL Chicago: McCutchan.

Spiegel, J. (1971). ItansactiensteJnterpiaiLbeMeeandixiduaLiamily.
anosociojy. New York: Science House.

Stott, D. H. (1966). Studiesefltoublesemecnildren. London: Tavistock.

Summers, A. A., 8 Wolfe, B. (1975, February). Which school resources help
learning? Efficiency and equity in Philadelphia public schools.
B . B . .

Tennent, T. G. (1971). School non-attendance and delinquency. Educational
Research. 1.3. 185-190.

Toler, C. (1974). The personal values of alcoholics and addicts. Newsleitei:
cLBesidentiaLMenialelealtnendBehayieraLScience.15(3),17-20.

Tennies, F. (1957). Communityandeeciety. New Brunswick: Transaction

Books.

Trueba, H. T., 8 Delgado, Gaitan, C. (1985). Socialization of Mexican
children for cooperation and competition. Sharing and copying. Journal

etEducaiIenaLEcmtiLandieadeEshm. 5(3),189-204.
Tyerman, M. J. (1968). Iniancy. London: University of London Press.

180

Villegas, A. M. (1988). School failure and cultural mismatch: Another view.
UrbarLBexiew. 29(4). 253-265.

von Mering, O. (1961). AgrammaLoLhiimanyaiues. Pittsburgh: University
of Pittsburgh Press.

Walberg, H. J., Bakalis, M. J., Bast, J. L., 8 Beer, S. (1988). We
euLchildEen. Chicago: Heartland Institute.

Waller, D., 8 Eisenberg, L. (1980). School refusal in childhood: Apsychiatric-
paediatric perspective. In L. Hersov 8|. Berg (Eds..) QuietschcelL

Modernperseectmesflimancyandechoefieflrsal (pp. 209-230). New
York: John Wiley 8 Sons.

Wetzel, J. R. (1987). W. Washington,
DC: Youth and America’s Future, the William T. Grant Foundation

Commission on Work, Family and Citizenship.

Wherly, B., 8Nelson, W. (1986). Iheassimiiaiienjndacculturatiencj

Indechineserefugeesjmefliincisjcnoels. (ERIC Document
Reproduction Service No. ED 278 741)

MICHIGAN STATE UNIV. LIBRARIES
IIIIIIIIIII]IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
31293010204885